CHAPTER 5
COMPUTER-AIDED
LEARNING
AND USE OF THE
INTERNET
Mark Huckvale
University College London, United Kingdom
Christian Benoit
Institut de la Communication Parlée, France
Chris Bowerman
University of Sunderland, United Kingdom
Anders Eriksson
Umeå University, Sweden
Mike Rosner
University of Malta, Malta
Mark Tatham
University of Essex, United Kingdom
Computer-Aided Learning and Use of the Internet
Briony Williams
University of Edinburgh, United Kingdom
1
Introduction
The working group in Computer Aided Learning and Use of the Internet
aims:
•
to bring together information about resources and tools for
computer aided learning (CAL) in the field,
•
to promote the use of the internet to disseminate information and
teaching materials,
•
to make suggestions of areas of resource development to potential
authors,
•
to encourage the generation of a comprehensive set of high quality
resources, and
•
to create an infrastructure for the management of resources.
The working group consists of educational professionals who have direct
experience
of
applying educational technology within Speech
Communication Sciences. Through our research into the current use of
computer-based methods, and the identification of successful
applications, we hope to encourage others to experiment with their use.
In the following sections we look at the match between Internet
technology and educational requirements in general, highlight the
requirements of Speech Sciences in particular, give a taxonomy of
available resources (with examples), criticise the current position, and
discuss what are the most urgent requirements for computer-based
teaching materials in Speech Communication Sciences. This review is
based on a database of 80 web-based resources (see Appendix I).
2
Technology
In this section we look at the tools available for CAL on the Internet and
World-Wide Web (we use ‘the Web’ to indicate the World-Wide Web and
Internet together). We look at telematic learning in general (learning at
a distance using technology) and CAL tools for Web learning
environments.
2.1
Telematic Computer Aided Learning
Interest in telematic education stems from what it can offer, namely,
deep learning. As adults learn well in social settings (Collis, 1995) which
promote sharing knowledge and resolving the resulting cognitive conflict
(Kaye, 1992), technologies which support such interactions are likely to
improve adult learning. Adults often have valuable personal experiences
94
Computer-Aided Learning and the Use of the
Internet
which they can bring to their learning, they need collaborative skills at
work, and they will need to develop independent learning skills with
peers when their formal education is over. These are some of the
reasons why telematic education has a significant part to play in
education.
There are two types of learning situation that we might wish to emulate
in telematic tutoring: static lectures and interactive tutorial/seminar
sessions. In order to be able to teach telematically, we need tools which
facilitate computer-supported co-operative learning (CSCL). CSCL classes
may be compared with contact classes along the following dimensions
(Palmer & Fields, 1994):
• synchronicity: students participate in the same session at the same
time (synchronous), or in sessions at different times (asynchronous).
• geogr aphi cal di st r i but i on: learners may or may not be at a single
site.
• homogeneity: learners may or may not all be similar in kind.
• coupling: the students may or may have to work fairly independently
(loosely coupled) or in a team (tightly coupled).
Tools which support synchronous activities are generally used to realise
a learning environment similar to a traditional classroom. However, this
technique is not suited to situations where very large numbers of
participants conduct multi-way discussions across different time zones.
This is a common situation on distributed web-based courses. Hence
asynchronous techniques have been developed. A typical asynchronous
technique uses a form of notice-board with the difference that, instead of
pinning up notices, participants attach computer files which may contain
text, graphics, video etc. Over a period of weeks, it is possible for
participants to respond to others’ files and form chains of discussions.
The key differences between more traditional knowledge-delivery
education and a ‘student apprenticeship’ style, more suited to telematic
tutoring, are outlined in table 5.1. Ideally, telematic tutoring is active
and discovery-based, and casts the tutor in the role of a facilitator of
learning. Also the learning materials used on the Web are globally
accessible. Hence it is important to keep materials up-to-date, in order to
encourage maximum use of the materials at a given site. This is why
authoring tools are important.
Table 5.1: Old and new forms of learning
Knowledge Delivery Education
Lectures
Student Apprenticeship
Discovery
Passive, solitary
Active, team-based
Omniscient tutor as controlling
Facilitative tutor as equal resource
gatekeeper
95
Computer-Aided Learning and Use of the Internet
Homogeneous
Diverse
Stable content
Changing content (tools used)
2.2
Internet Technology
The Internet is an information carrier - a medium - which links
information providers to information users. We are concerned with the
pedagogic potential of this medium: its strengths and weaknesses in
respect to educational provision and educational requirements. Here the
Internet is taken to mean both the physical hardware which
interconnects millions of computers, and the software tools which can be
used to enable teaching and learning to take place world-wide. Few of
the tools were originally designed for pedagogic purposes (many arose
from efforts to enable Computer-Supported Co-operative Work and
Computer-mediated Communication) but many have been successfully
used to enable teaching and learning over the internet.
There are many tools available for general authoring of telematic
tutoring materials. The key general tools that can be used to prepare
materials are shown in table 5.2, under the headings of Authoring
(generating teaching materials), Components (the telematic tools
themselves) and Tracking (tools to handle registration, payment and
assessment).
Table 5.2: Tools for authoring Internet teaching materials
Tool
Det ails
Authoring Tools
•
•
•
•
Authorware: CAL production
HotMetal or MS FrontPage
•
HTML writing tool
MS-Office 97
•
Paper and WWW authoring tools
AC3D
•
3D VRML authoring tool for Suns
ActiveX Control Pad
•
HTML writing tool with support for
Authorware & Director with
Photoshop/Freehand
and Shockwave plugins
+ JavaScript/VBScript
Director: movies
Photoshop/Freehand: graphics
Shockwave plugin: WWW browsing
JavaScript and VBScript authoring
•
Also: ActiveX applications
Telematic Components
Computer conferences/Newsgroups:
•
(NNTP, Notes, 1st Class)
Enables discussion forums to be set
up and administered
asynchronously
Videoconferences (JPEG, ITU H.320)
•
Synchronous videoconferences
Chat: IRC, MS comic chat
•
Synchronous one-to-one or
many-to-many discussion
Shared applications and whiteboards
•
Cooperative working on whiteboard
or other application package
96
Computer-Aided Learning and the Use of the
Internet
Email (MIME)
•
Asynchronous file transmission
Video on demand (Sun ShowMe)
•
Asynchronous videoconferencing
and materials distribution
Search engines
WWW
browser:
Netscape/Explorer
•
Locates material on WWW site
•
Views material
(HTML, Java, Java/VBScript, ActiveX)
Tracking Tools
WEST/TopClass
•
Unix/WinNT-based student
tracking tools
WebCT
•
U nix-based student tracking tools
CAFE-MONDIAL
•
WinNT-based student tracking tools
Authoring CAL for the Web or for standalone use takes much effort. A
package whose use in class might last from 30 minutes to 3 hours might
take between 4-6 person-months of effort, particularly for high-quality
multimedia materials with a good look and feel. There is a need for
professional graphic design input, video production, CAL and pedagogical
input. This makes commercial-quality CAL production a professional
activity. Nevertheless, there are a number of tools available that enable
the production of useable courseware in a shorter time. There are two
main types: Web-only tools and Web/local tools.
There are many Web-only tools available, of which MS FrontPage and
HotMetal currently receive the most accolades. MS Office 97 enables all
the MS Office tools to save in Web format for direct publication. This is a
good initial solution for rapid prototyping of materials (ie. documents,
presentations, databases and spreadsheets) so that they can be quickly
available on the Web. However, tools allowing greater graphical and
animation sophistication are also required. Those not using a Microsoft
server will need to know some HTML and scripting to achieve the same
results. This is where the MS ActiveX control pad can offer some
rudimentary support.
Of those tools which are useable both in standalone mode on a local
machine and also on the Web, Macromedia’s authoring suite is a
favourite. It enables good-quality CAL to be produced and, with the
appropriate Shockwave browser plugin, allows it to be incorporated into a
Web page. This solution enables flexible software to be produced using a
productive, high-level development tool that does not require knowledge
of programming.
A variety of tools are detailed below with an indication of their typical
use (lectures or tutorials). Lectures can easily be given in
videoconferencing systems. There are a variety of these from shareware
Internet systems (CuSeeMe) which are low -quality, through to ISDN tools
(e.g. PictureTel) which need specialised hardware (approximately 600 ecu
for a PC). Lectures may also be delivered as files on a Web page, for the
97
Computer-Aided Learning and Use of the Internet
student to work through, and these may be viewed using a browser. More
interactive, tutorial-like sessions may be delivered using computer
conferences (e.g., 1st Class, newsgroups) which enable users to send
text files to a named discussion group, and to build up a set of
interlinked messages in the style of a debate (email can be used to
similar effect but does not provide any ‘threading’ facilities to display how
messages are interlinked). Tutorial sessions can be provided by the
following methods, all of which provide a high degree of interactivity:
•
•
•
Web pages: students write their presentations and publish them on
the Web or in a conference.
Videoconferences.
Whiteboards (built into Netscape) and other applications shared
between all the participants.
Table 5.3 Use of telematic components
Computer
S
A
D H L
T
conferences/
y
s
i
o
o
i
newsgroups
n
y
s
m o
g
c
n
t
o
h
s
Videoconferences
h c
ü
g e t
ü ü ü ü
Chat: IRC,
ü
ü ü ü ü
ü
ü ü ü ü
MS comic chat
Shared
applications and
whiteboards
Email
ü ü ü ü ü
Video-on-demand
ü ü ü ü ü
Search engines
ü
ü ü ü ü
WWW browsers:
ü ü ü ü ü ü
Netscape/Explorer
Chat programs (such as Internet Relay Chat and MS Comic chat) can
provide synchronous discussions in text between multiple users. An
additional tool is video-on-demand. Using a Sun video-on-demand server,
lectures and tutorials can be videorecorded and sent out on demand to
students. Such technology requires good network connections and is
beyond the current bandwidth of the Internet (but is possible over private
or ISDN lines). Table 5.3 illustrates the types of use to which telematic
component tools can best be put.
Once we have produced a course and made use of the appropriate
component tools, we will wish to register telematic students and track
their progress. This is the domain of tracking tools. A number of tools are
98
Computer-Aided Learning and the Use of the
Internet
available: WEST (which is Unix-based), TopClass (an NT version of WEST),
WebCT (a Unix-based system) and Cafe-Mondial (an NT-based system).
The first three are currently commercially available. Cafe-Mondial is
under development as part of an EU Telematics programme. The tools
provide facilities to enable students to do the following:
• browse through a module catalogue,
• register,
• work through the teaching material,
• undergo assessment,
• inspect their marks, and
• claim credit at participating institutions.
2.3
Summary
An important character of Internet education is the potential for teaching
and learning at a distance from a host institution. The consequence of
this telematic teaching is that the students may well be unable to come
together synchronously as a whole class (for reasons of time or distance).
For such students, a key feature of the Internet is its ability to link them
together into a collaborating community by means of file-based computer
conferences, e-mail and the Web. They can benefit from a richer
teaching and learning environment. Furthermore, this environment can
include both static information and also active and interactive content in
full multimedia format, using Java, Visual Basic, Perl and other scripts
as well as ActiveX and Java mini-applications.
While the technology of the Internet is seductive, it is nevertheless
important that the technology be used to produce well-motivated,
pedagogically sound learning experiences which play to the strengths of
the medium. These clearly lie in the facility with which material can be
made available, maintained, updated and re-used. The sheer volume of
the material available, and the social dimension of community which
telematics creates, are strengths in themselves. After all, education is
concerned with preparing learners to become fully-fledged participants in
a given arena. Participation in education is not only a matter of acquiring
knowledge, but also involves acquiring skills in presentation,
communication, analysis, debate and discussion. It is in this area that
the Internet currently excels. It may not always be appropriate to publish
material on the Internet when a book or compact disc would be more
practical. However, the Internet is unrivalled for bringing together
scattered people in virtual communities for teaching and learning.
99
Computer-Aided Learning and Use of the Internet
3
Speech Sciences Education
3.1
General features of speech sciences education
There are certain features of education in the speech sciences which
make the Internet especially suitable as a medium for education While
conventional methods will still form the foundation of speech sciences
education, it is plain that the Internet can make a contribution that is to
be especially welcomed by this field. The particular features of speech
sciences education are outlined in the subsections below.
Multiple disciplines
The study of Speech Sciences is inherently multidisciplinary to an
unusual degree, and involves elements of linguistics, modern languages,
psychology, computer science and electrical engineering, as well as
many biomedical areas. Students find the multiplicity of disciplines
difficult to manage, since their background rarely prepares them in
breadth. No student comes fully grounded in all relevant areas.
Additionally, the field crosses the divide between arts and sciences.
From the teacher’s point of view, this situation complicates the
administration and delivery of education, since traditionally expertise in
these areas has been distributed across a number of academic
departments, each with its own human resources, facilities and tools.
Given such a lack of centralisation, the Internet as a medium of
communication can help in a number of ways:
• It gives students the means to draw on varied expertise throughout
the community, consulting outside tutors and experts.
• It facilitates communication between “virtual” communities of
teachers or students working in a particular sub-discipline.
• It enhances quality and reduces a duplication of effort in the creation
of teaching materials.
• It encourages the dissemination of good practice among teachers.
Practical skills
The Speech Sciences student needs to acquire practical skills in
listening, analysis and performance. In addition, speech and language
therapists need to acquire basic clinical skills. This inevitably requires a
lot of staff/student interaction at a time when there is pressure
throughout European universities to increase efficiency and decrease
the use of staff resources. Interactive assessment and peer assessment
through the Internet, together with interactive online tutorials, can
increase the efficiency of student self-study in these skills.
100
Computer-Aided Learning and the Use of the
Internet
Multiple media
Speech processing is inherently multimedia in nature, involving both
sound and vision. Current printed materials cannot convey the coordination of articulation, sound production and sound perception. The
Internet standards for audio, video and even three-dimensional models
could lead to materials where such interactions become much clearer to
the student (in the case of a 3D model of a vibrating larynx, for example).
Extramural interaction
Many students (in modern languages or computer science, for example)
find their interests in Speech Sciences cannot be satisfied by their local
institution.
There are relatively few specialised departments of
Phonetics or Speech Science in Europe. Even in larger departments,
postgraduate students might find themselves academically isolated
because of the specialist nature of their interest. The Internet can allow
students to participate at a distance with teachers and experts in other
institutions, using the variety of mechanisms for computer mediated
communication.
In the same way, students can form “virtual
communities” using the Internet, and these can be a valuable source of
support to students who are academically isolated within a small
department. In extreme cases, an extramural tutoring relationship can
be set up using the Internet, at least for part of a student’s course.
Finally, among teachers, the Internet can be a very efficient and costeffective way of sharing teaching materials and disseminating good
practice, to avoid duplication of effort.
Language diversity
Speech Science education should be built on both the universals and the
specifics of human languages. The Internet can allow easier access to
linguistic resources, tutorial information and native speakers of various
human languages, regardless of geographical location. It is true that
Web materials are overwhelmingly written in English. However, the
Internet also provides the easiest manner of contacting and interacting
with native speakers of languages other than English. This would be
useful in the case of experiments as part of a class project in phonetics
or phonology.
3.2
The Internet versus CD-ROM
Certain of the teaching functions indicated above could equally be
provided by a CD-ROM, especially one that was updated frequently.
Other functions, however, necessarily require the Internet.
The
following subsections outline which teaching functions belong in which
category.
101
Computer-Aided Learning and Use of the Internet
Teaching resources that could be provided on CD-ROM
CD-ROMs have the advantage of much faster access speeds than the
Internet, and less risk of a temporary denial of service due to network
failure. There is also less of a virus threat, since the manufacture of a
CD-ROM can be closely controlled. Also, sending out CD-ROMs means
that the originator’s computers are not at risk from unauthorised access.
The following teaching functions could be successfully implemented
using CD-ROMs.
• Intensive training in practical phonetics. This is a very clear case
where computer-aided learning would be useful. Students require
intensive practice which traditionally has been facilitated by specialist
phoneticians, at great cost in staff time. Much of this practice could
be done by students working with self-study modules. Staff input
would be required only occasionally, to correct and check the
individual student’s efforts.
• Mul t i di sci pl i nar y t ut or i al mat er i al s. This is another clear case of
the advantages of computer-aided learning. Tutorial materials on CDROM could draw together elements from several different fields which
may otherwise not be easily accessible (e.g. physics and anatomy, in a
module on the perception of sound).
• Mul t i medi a pr esent at i on of mat er i al s. While a few multidisciplinary
textbooks exist, CD-ROMs have the advantage that sound and
animation can be added in order to make the principles clearer to the
student. Figure 5.1 is an example of what can be done using CD-ROM.
It is taken from the demonstration version of the “Sound” module of
the PsyCLE software, which was funded by the UK Higher Education
Funding Councils’ “Teaching and Learning Technology Programme”.
Further details are available at [URL1].
Figure 5.1: This screenshot has animation at two places: the panel showing
alternating waves of compression and rarefaction, tied to a rotating object (with
degrees of arc indication, changing synchronously); and the sinusoidal waveform in
the display panel underneath, which changes in synchrony with the other display.
102
Computer-Aided Learning and the Use of the
Internet
Teaching resources for which the Internet is needed
• Access t o wor l d exper t s. If a student (especially a postgraduate
student) wishes to consult an expert about a question for which no
expertise is available locally, then the Internet is much the fastest
(and cheapest) method of doing so.
• Access to fellow-st udent s. For an academically isolated student, the
Internet provides access to a “virtual learning community” of other
students of the same subject, for mutual support.
• Access to native speakers. If a student seeks native speakers of
another language for a project, then it may be possible to use the
Internet to contact them, much more easily and cheaply than would be
possible by traditional means. However, recording native speaker
utterances over the Internet may not yield sufficiently good sound
quality for phonetics experiments.
• Access to research on foreign languages.
A large amount of
reference material is available over the Internet. In the case of
foreign languages, it may be difficult for the student to find the
desired information locally, and this is where the Internet could be of
use.
• Access t o t ut or s. For students unable or unwilling to attend higher
education institutions, the Internet can provide a convenient means of
access to tutors.
• Videoconferencing and transmission of recorded lectures.
An
obvious use for the Internet is in videoconferencing between students
(such as when engaging in a “virtual tutorial”), or in the multicasting
of recorded lectures. At the moment the latter requires specialist
high-speed connections and equipment.
• Access to fellow-t ut or s t o shar e good pr act i ce. Among university
teachers, the Internet can be used to share teaching materials and to
disseminate good practice. This is already being done, in the UK, by
Government-funded initiatives such as the “NETPhon” network.
Current bottlenecks in Internet provision
Many of the possibilities mentioned above cannot come to full fruition
until the technology matures sufficiently to implement them, or until
existing technology becomes much more widespread and accessible.
There is also, naturally, a question of funding, as much of the requisite
technology is expensive. Although computer-aided learning will save on
staff costs in the long term, the initial costs are a barrier to full
103
Computer-Aided Learning and Use of the Internet
implementation in some cases. The subsections below outline some
current technical barriers to provision.
• Limited access to MBONE technology: The MBONE is a technology
for broadcasting on the Internet suitable for viewing transmitted
lectures. Since it is not yet widely available, very little has actually
been made of its potential.
• Li mi t ed bandwi dt h: Even using the higher capacity portions of the
network (the UK Universities high-speed network, SuperJANET is
capable of 34 megabits per second) the speed of transmission is not
sufficient to support some of the applications outlined above. In view of
this, the Scottish Higher Education Funding Council has recently
funded four “Metropolitan Area Networks” capable of 155 megabits per
second. These four interconnected MANs cover most Higher Education
Institutions in Scotland, and use the new Asynchronous Transfer
Mode (ATM) technology.
• Real-time audio : This is currently available using client/server
architecture and proprietary compression formats (e.g. RealAudio from
Progressive Networks: details at [URL2]). However, the use of this
technology is not yet widespread. Since whole radio programmes and
lectures can be made available for listening over the Internet using
this technology, it is an important building-block for the educational
use of the Internet.
• Real-time video : There is also a need for real-time video using
compression software, for use over existing connections (e.g.
RealPlayer from Progressive Networks). This technology is currently
adequate for lectures etc, but possibly not for detailed phonetic
demonstrations, where greater visual clarity is required.
Support programmes for Computer-Aided Learning
There are various support programmes and networks available,
particularly in the UK, for Computer-Aided Learning. These are in two
main categories: general CAL programmes (supporting the development
and use of CD-ROM material), and programmes that focus on the use of
the Internet. The principal UK programmes are outlined below; we hope
to collect similar information from other European countries in future
years.
General CAL support programmes for higher education
•
Teaching and Learning Technology Programme (TLTP):
This
programme is funded by the UK Higher Education Funding Councils to
support the development (by staff consortia or by single institutions) of
104
Computer-Aided Learning and the Use of the
Internet
a large amount of software available at minimal cost to UK HE
institutions. Details are available at [URL3].
•
Comput er s i n T eachi ng Ini t i at i ve (CTI): This programme is funded
by the UK Higher Education Funding Councils to facilitate effective
use of computers in Higher Education through the provision of
information on technology to support educational change. Details are
available at [URL4].
•
Learning Technology Dissemination Initiative (LTDI):
This
programme is funded by the Scottish Higher Education Funding
Council to support and encourage academic staff in the integration of
technology into teaching and learning. The focus is on delivering
material to where it can be used. Details can be found at [URL5].
•
Teaching and Learning Technology Support Network :
This
network evolved out of the Teaching and Learning Technology Project.
It supports institutions in all aspects of institutional policy relating to
the use of learning technology, and thus serves to complement the
more “classroom-based” expertise of the Learning Technology
Dissemination Initiative. Details are available at [URL6].
Support programmes focussed on the use of the Internet
•
UMI: Use of MANs Ini t i at i ve: This is a series of projects funded by
the Scottish Higher Education Funding Council to develop educational
applications which exploit the speed and bandwidth available on the
Scottish high-speed Metropolitan Area Networks.
Details are
available at [URL7].
•
T ALi SMAN: T eachi ng And Lear ni ng i n Scot t i sh Met r opol i t an Ar ea
Net wor k s:
This project is funded through the Scottish Higher
Education Funding Council’s “Use of MANs” Initiative. It aims to
deliver high-quality training on the most effective use of MAN
technology to the academic community in all Scottish HE institutions.
Details are available at [URL8].
•
NETPhon: Network for education and training in phonetics. This
was established in 1994 and initially funded by the Department for
Education and Employment.
The network held meetings and
sponsored the electronic discussion list “Phonet”.
Details of its
activities are available at [URL9].
•
eLi b: t he El ect r oni c Li br ar ies Programme. This is funded by the
Joint Information Systems Committee of the UK Higher Education
Funding Councils.
It aims to pave the way towards the fully
electronic academic library of the future. Details are available at
[URL10].
105
Computer-Aided Learning and Use of the Internet
•
Net sk i l l s: Net wor k sk i l l for t he UK Hi gher Educat i on Communi t y.
This programme aims to help the UK Higher Education community to
make effective use of the Internet for teaching, research and
administration. It is part of the Electronic Libraries Programme
(eLib). Details are available at [URL11].
3.3 Prototypes of Internet-mediated speech sciences education
The subsections below give examples of Web pages which offer working
prototypes for what can be expected in Web-based speech sciences
education. Each field is considered separately: phonetics/phonology,
spoken language engineering, and speech/language therapy.
Phonetics and Phonology
Education in phonetics and phonology involves such practical skills as
the phonemic transcription of utterances, and the ability to interpret the
information in a sound-spectrogram. Both these skills require extensive
practice, which is expensive in terms of staff time, and this is where
computer-aided learning can be of particular assistance.
Figure 5.2: This screenshot (above) is an example of a Web page with online
transcription practice and self-testing [URL12]. The first column gives words for
transcription, with the chance to hear the recording. The second is for students’
106
Computer-Aided Learning and the Use of the
Internet
transcriptions. The third provides phonetic symbols, and the fourth column, when
clicked on, shows the correct answer.
Figure 5.3: This screenshot (above) is a “mystery spectrogram”, where the student
must read the spectrogram and determine the utterance [URL13]. The first
successful student is given credit by placing his/her name on the page together with
the correct answer, and a new spectrogram is then provided. This page adds the
incentive of a small competitive element to this classic training technique in acoustic
phonetics.
107
Computer-Aided Learning and Use of the Internet
Spoken Language Engineering
Education in spoken language engineering demands a detailed
understanding of acoustics and relevant applied mathematics, together
with a practical understanding of speech technology applications. The
screenshots below are examples of the possibilities ava ilable in these
areas, both interactive and non-interactive.
Figure 5.4: The simplest application of Web technology is the provision of text
material, such as lecture notes, in hypertext form. This screenshot (above) shows a
particularly detailed set of notes in the field of speech analysis [URL14]. This mode of
publication might form a component of a “virtual university” where most or all of the
teaching is conducted over the Internet. There will always be a place for noninteractive reference material of this kind in any Web-based course.
108
Computer-Aided Learning and the Use of the
Internet
Figure 5.5: This Web page allows the user to type in arbitrary English text and
submit it to a text-to-speech synthesiser at AT&T Research [URL15]. The
synthesised speech waveform is then sent to the user immediately, and can be
listened to if the appropriate hardware is installed. The user has a choice of several
pre-set “voices”, such as ‘male’, ‘female’, ‘child’, ‘raspy’, ‘gnat’. However, what is of
more interest from the point of view of education is the fact that the user can alter the
main parameters of the voice and listen to the result. This will allow the student to
gain practical insight into the effect of these parameters on a synthesised voice. This
is a vital prerequisite to advanced education in speech technology.
109
Computer-Aided Learning and Use of the Internet
Speech and Language Therapy
The field of speech and language therapy is not yet as far ahead as the
other two fields in its use of the Internet. Therefore the exampl es below
are taken from a different (but related) field, as an example of the kind of
Internet-based education appropriate for the development of clinical
skills in students at an early stage of their professional training.
Figure 5.6: The “Interactive Patient” has been developed at Marshall University
School of Medicine [URL16]. It is a study of a consultation with a patient with acute
backache. The student first asks questions to learn the history of the complaint, with
answers made available at once. On a later page, the student “examines” parts of
the patient’s body by clicking on images of it, with the results returned at once. The
110
Computer-Aided Learning and the Use of the
Internet
student can also specify certain laboratory tests and X-rays, the results of which are
made available in detail.
Figure 5.7: Once students are confident that they can diagnose the condition, they
choose from a menu and then prescribe treatment from another menu. Their choices
are emailed to the site, which will email the student with a note of how correct their
answers were. This kind of software could be adapted to simulate a clinical
encounter in speech therapy, for example the measurement of a child’s hearing. In
this way, novice students would be enabled to learn correct procedures in an
inexpensive way, before practising on real subjects. Another advantage of this
simulation is that it is not necessary for the students to have access to expensive
specialised hardware for use in practising their clinical skills. Thus it could be used
as part of a distance learning module over the Internet.
111
Computer-Aided Learning and Use of the Internet
3.4
The shape of things to come?
An even more radical development would be the use of the Internet as
the primary medium of education.
The following screenshots are
examples of two fledgling projects with this aim. The Clyde Virtual
University focusses mainly on proving the technology in a few prototype
fields (such as a Web interface to student newsgroups for email
conversations sorted by topic). On the other hand, the University of the
Highlands and Islands aims to establish a new university, focussing on
the actual delivery of education rather than simply testing software.
Figure 5.8: Clyde Virtual University is one of the projects funded by the SHEFC’s
“Use of MANs Initiative” (see 3.2 above). Its aim is to develop the software necessary
for a virtual university, and to evaluate it in use by providing prototype applications for
use by student subjects. Several different kinds of software must be integrated, and
112
Computer-Aided Learning and the Use of the
Internet
this screenshot indicates the various functions that are expected of a virtual
university: courseware and tutorials, assessment, reference material, inter-student
interaction, and student registration and tracking. This page is at [URL17].
Figure 5.9: The University of the Highlands and Islands is an embryonic project
composed of several further education colleges in Scotland. It is the first real attempt
to create a virtual university, and at the moment it is in the preparation stage. It is
closely tied to local needs and employment opportunities, but hopes eventually to
include a wide range of academic subjects. It is made possible partly by the rare
combination of a very low population density (in the north of Scotland and offshore
113
Computer-Aided Learning and Use of the Internet
islands), a very high level of educational and technological development (in the UK as
a whole), and the installation of high-speed telecommunications links (in this area of
Scotland). The UHI is funded partly by the UK’s Millennium Commission, and partly
by local authorities. This Web page is at [URL18].
4
Survey of Internet Resources
Between February and June 1997, the working group surveyed Web sites
world-wide for educational resources relevant to Speech Sciences. A
W eb-based questionnaire was used to collect details [URL19].
The
contents of the inventory are published on the Web on a regular basis. It
can be viewed at [URL20].
From a study of the range of resources in the inventory, the group has
developed a set of nine categories to describe the types of educational
resources currently available on the Internet. In the sections below, we
give a brief description of each category and describe one or two
resources to give more indication of the type of material available. A
more complete list of resources found under these headings is given in
the Appendix.
4.1
General information
In this category are documents relating to how the Internet may be used
to deliver teaching and learning.
These cover the pedagogical
opportunities and problems arising from the use of the internet, to
discussions of the individual technological services.
An example is the paper “Delivering Instruction on the World Wide Web”
by Thomas Fox McManus from the University of Texas at Austin [URL21].
In this paper McManus discusses the advantages and disadvantages for
using the Web for instruction delivery, and reviews the technological
support for authoring materials. In addition McManus also discusses
how theories of instruction could play a part in the design of the
materials. In the instructional systems design model, the materials are
designed in terms of a set of desired ‘behaviours’ expected of learners
once they have completed the course. In the Cognitive Flexibility model,
materials are organised into a web of instructional sequences that are
anything but linear, to accommodate the different requirements of
individual learners. This hypermedia delivery is, of course, well suited
to the technology of the World Wide Web.
In a related paper, “Some considerations for Designing Internet Based
Instruction”, [URL22], McManus makes the comment:
“The internet can deliver video, but not as quickly as videotape,
television or CD-ROM.
It can carry real-time personal
114
Computer-Aided Learning and the Use of the
Internet
interaction, but not as well as telephone or video conferencing.
It can display textual information, but not as usefully as a book
or magazine.”
Why then should the Internet be useful for educational materials?
McManus replies:
“It combines advantages of other media so that it conveys
video and sound better than a book, is more interactive than
videotape, and unlike CD-ROM, it can link people from around
the world cheaply. The second advantage, and one that is
often overlooked when discussing the Internet as a delivery
system, is that it can also be a content provider.”
4.2
General Tools
In this category are Internet accessible tools to aid in the delivery of
tuition: both in terms of computer-based and internet-based
technologies.
This ranges from tools for authoring and interactive
assessment to complete environments for the creation of telematic
courses.
An example is the WebCT package [URL23] that covers student tracking,
assessment, chat, discussion groups and e-mail as well as course
material. Although not many course development tools of this kind exist
at present it is likely that any such tool will be similar to WebCT in terms
of general structure, although details may be varied infinitely.
WebCT is a tool for creating a web-based educational environment. It may
be used to create complete on-line courses or supplements to existing
courses. Interaction with WebCT is through any of the existing web
browsers and all content is in the form of HTML documents. The server
on which WebCT runs must, however, be running the UNIX operating
system. A wide variety of UNIX dialects are supported.
A course is organised around a home page with entry points to other
components (exercises, chat, notice boards, quizzes etc.). There are four
levels of users: administrator, designer, marker, and student. The
administrator sets up the system, the designer constructs a specific
course and, markers administer tests and quizzes. The system contains
tools for student tracking and a simple tool for presenting results in the
form of histograms. Any course material, experiment etc. which is either
in the form of a HTML document or a program written in the Java
programming language may be part of WebCT. Material in any other form
necessitates leaving WebCT. A course on acoustic phonetics, for example,
would require students to work with acoustic analyses outside We bCT. In
115
Computer-Aided Learning and Use of the Internet
such a case the web course would serve as a complement to other forms
of teaching material. All this may change, of course, if new programs for
acoustic analysis etc. are developed in the Java language, but at present
such programs do not exist.
It is estimated that the program will be released in August 1997. At the
time of writing (June 1997), the tool was not available as a commercial
product, but only accessible to those accepted as beta testers.
4.3
Tutorials
This category is for educational materials relevant to the Speech
Sciences theme, which are directed at the student. They can cover both
large and small topics, from complete tutorials in Phonetics and
Phonology, to a discussion of the principles of spectrography.
An example is a tutorial on acoustic-phonetics [URL24] developed by H.G.
Tillmann and F. Schiel of the Ludwig-Maximilians-University in Munich.
The tutorial is basically a text-based course, divided into four chapters.
Each chapter ends with a few questions on the topics covered in text.
Suggested answers are linked to the question sections. From this point
of view the tutorial is not much different from a written text. The tutorial
makes use, however, of one of the advantages of electronic media – the
possibility of linking non-text material to the text as illustrations. Thus
there are several sound clips linked to the text, illustrating various
aspects of sound qualities. The reader may listen to these sounds by
simply clicking on a button as illustrated in Figure 5.10.
Figure 5.10. One of the advantages of hypertext tutorials is the possibility of linking
various non-text material to the text. The Figure shows a diagram of a complex wave
form. By clicking on the button to the right, the reader may listen to the sound
produced by this wave form. The illustration is copied from a tutorial by H.G. Tillmann
and F. Schiel of the Ludwig-Maximilians-University in Munich [URL24].
The possibility of linking sounds to the text is, of course, of particular
interest in the field of phonetics. This is true for acoustic phonetics as
illustrated above, but also for phonology, foreign language acquisition and
for many other topics. Several tutorials linked to the data base use this
option. A further example is a tutorial on auditory perception [URL25]
which is richly illustrated with sound clips.
116
Computer-Aided Learning and the Use of the
Internet
Using the Java language capabilities, it would be possible to include
actual experiments in the text. However, no tutorial that the authors
have come across utilises this option. It is also possible to include
animated graphics and video clips. This would be particularly useful in
connection with texts on articulatory phonetics and aspects of non-verbal
communication, and bi-modal communication including visual and
auditory cues. A demonstration of audio-visual speech synthesis [URL26]
may serve as an example, although it is not primarily intended as a
tutorial, but as a description of research work on audio-visual speech
synthesis conducted at the ICP institute of the University of Grenoble. By
clicking on links, the reader may load video clips illustrating various
aspects of audio-visual synthesis. The synthetic head used in some of
the demonstrations is shown in Figure 5.11.
Figure 5.11: The figure shows a wire mesh model of a human face and the synthetic
face, based on the model, used in the audio-visual speech synthesis project
conducted at the ICP institute at the University of Grenoble [URL26].
The usefulness of video clips and other than short sound clips is at
present somewhat limited by bandwidth. The file size of the video
illustrating the ‘talking head’ mentioned above is about 1.5 Mb. On slow
networks this may be a severely limiting factor.
4.4
Component Resources
In this category are smaller educational resources that could be included
in tutorial material, but which are not currently suited to student self
study.
A typical example is a recreation of the sights and sounds of the Haskins
pattern playback machine [URL27]. The now classic pattern playback
system was developed in the fifties at Haskins laboratories. The Web
pages contain a fairly detailed description of the system and several
sound clips to illustrate the sound quality of the synthesis. This
component may be a valuable illustration in any course on speech
117
Computer-Aided Learning and Use of the Internet
synthesis, but also as an illustration of a technique which has played an
important role in many classical perception experiments.
Another example [URL28], also from Haskins, illustrates the principles of
articulatory speech synthesis. The different steps in the synthesis of a
set of vowels are shown – vocal tract shape, area grid, area function,
spectrum of the sound. The resulting sound may be listened to by
clicking on a button. At present, the system can only produce static
vowels, but work is in progress to enable the synthesiser to simulate
dynamic processes as well. Diagrams showing two of the synthesis steps
are shown in Figure 5.12
As was mentioned above, use of the Java language makes it possible to
run actual experiments via a web browser. Programs written in the Java
language may be loaded and run using a Java enabled browser. An
example of this kind of usage of the web is a demonstration of categorical
perception produced by one of the authors [URL29].
Figure 5.12: Two diagrams illustrating steps in articulatory speech synthesis. The left
panel shows a vocal tract with grid lines used in calculating the area function, which is
shown to the right.
A categorical perception test consists of two parts – a categorisation test
and a discrimination test. If perception is to be classified as categorical it
is necessary for subjects to classify stimuli chosen along a continuum
into a limited number of categories, normally two or three. It is further
required that discrimination should be considerably better across
categories than between them. The demonstration is a complete test for
categorical perception. After each part of the test, the result is presented
in the form of a diagram (see Figure 5.13).
118
Computer-Aided Learning and the Use of the
Internet
Figure 5.13: The figure shows a diagram illustrating the result of the classification
test, which is part of a categorical perception test.
It is a demonstration rather than a scientific experiment only in the
sense that in an actual experiment one would require the subjects to
listen to a greater number of stimuli. There is, however, no reason why
this technique could not be used also to perform real experiments.
As far as the authors are aware, this demonstration is the only resource
of its kind, but it seems likely that with the development of the Java
language and subsequent wider use of it, one may see many more
programs of this type in future. One obvious advantage is the relative
platform independence. Browsers which run Java applications exist today
for all major platforms.
The examples described here may be used as components in relevant
courses. The advantage of components over complete courses is, of
course, that they permit teachers to compose their own syllabi freely and
only use what components they see fit for a particular course.
4.5
Course Syllabus Information
This category includes descriptions of the syllabus or curriculum for
courses in the field of Speech Communication Sciences.
At their
simplest, these ‘course home pages’ contain lecture lists, reading lists
and assessment details; at their most complex, they include interactive
activities and links to related materials. Eventually, these will develop
into complete automated distance learning or ‘telematic’ courses. In
their current form they may be of more interest to teachers than to
students of other institutions.
A good starting point on the Web for these materials is the ‘World Lecture
Hall’, a hyperlinks page of references to courses world-wide that are
using the Web to some degree to deliver class materials. While the
material seems to be almost entirely in English and with a mostly U.S.A.
representation, the reference page for Linguistics [URL30] contains
useful examples of the range of Web materials currently used in this
119
Computer-Aided Learning and Use of the Internet
way. A course on the Human Instinct for Language investigates the
‘innateness hypothesis’; a course on the French language provides the
equivalent of introductory textbook in French linguistics; a course in
Literary Stylistics is based in Singapore; a course in Computational
Linguistics contains compl ete lecture notes.
We look in more detail at the course materials provided by Robert Beard
of Bucknell University in the U.S. called ‘Linguistics 105 - Sounds and
Words’ [URL31] since it contains some phonetics/phonology components.
The self-styled ‘on-line interactive syllabus’ contains
•
•
•
•
•
•
•
•
•
•
•
•
4.6
Course objectives
Course management and organisation
Course grading
Course reading
Course schedule with lecture dates, titles, readings and
assignments.
Glossary with self-study testing
Linguistics lexicon
Phonetics fonts
Links to on-line grammars
Links to pictures of saggital cross-sections of various
articulations
On-line quizzes on consonants, vowels and saggital sections
Videos on chimpanzee language, wild children, the origins of
language
Bibliographies and Refereed Articles
This category refers to on-line bibliographies and published articles.
Advanced students need access to descriptions of current research of
ensured quality. Thus this category relates to the access to published
papers from accredited sources.
Scientific Journals are more and more concerned with electronic access,
text retrieval, and maintenance of digital archives. Also book editors
think increasingly to make their book available in a digital form, either
on the internet or on CD-ROM. In addition, most of today big conference
offer either printed proceedings and/or CD-ROM; and some conference
proceedings are partly or entirely available on the network. Below, we
give some examples of what is available today in the area of Speech
Sciences and Technology.
A few books are already entirely available in a browsable format on the
internet, e.g., Survey of the State of the Art in Human Language Technology,
120
Computer-Aided Learning and the Use of the
Internet
R. Cole et al., Eds. [URL32], or the EAGLES guidelines, N. Calzolari & J.
McNaught, Eds [URL33]. Beside the fact that these books are free of
charge and available very quickly after final editing, they offer the
incomparable facility of cross-references as links to other sites on the
network. Unfortunately, such initiatives are still infrequent. However, it
is getting more common to find entire doctorate theses on the network,
even before they are defended in some cases.
Today, many books are published with accompanying CD-ROMs on which
the articles are often retrievable. More interestingly, these CD-ROMs
often include audio files or even movie files that cannot be published “on
paper”, for example: Progress in Speech Synthesis, J. van Santen et al.,
Eds, Springer-Verlag, 1996. It is also more common for technology books
to have an associated Web site where the reader can get updated
information after the publication.
Similarly, conference proceedings now appear more and more on CDROM. The objective here is twofold: through CD-ROMs, organisers of “big
conferences“ can produce proceedings more cheaply, and CD-ROMs are a
more manageable medium of communication for a thousand 4-page
articles. In this case too, it often happens that articles contain links to
the authors’ homepages, thus allowing newer data to be accessible. In
Speech Sciences, the proceedings of the ICSLP’96 (Philadelphia, USA)
and of the Eurospeech’97 (Rhodes, Greece) are published in both formats.
Also, some organisers make proceedings available to the public through
the internet even before the conference itself, thus allowing the
participants to come ready to the conference after having read the papers
(e.g., the ESCA Workshop on ‘Audio-Visual Speech Processing’, Rhodes,
Greece, Sept. 97; [URL34]).
There are basically three approaches to “electronic publication“ of
scientific articles in academic journals.
First is the digital storage of archives. In that vein, the ‘Journal of the
Acoustical Society of America’, for instance, now offers CDROMs where
(even very) old articles have been scanned and stored. This is certainly
a rather cheap and convenient way to get the entire collection of past
issues of the JASA. However, the texts are simply scanned so that text
retrieval is impossible on such CDROMs.
Second, electronic files may be accessible through the internet to help
understanding or to illustrate articles which are published in a paper
format. The URL addresses of accompanying files, whether audio or
movie files, now appear in the Speech Communication Journal since
beginning of 1996. Those files are publicly available from [URL35]. Also,
121
Computer-Aided Learning and Use of the Internet
a searchable table of contents is available under condition of subscription
to the Journal.
Third, a few journals are based on a complete and free access through
the web. This is the case of two relatively recent initiatives, one by ESCA
and ELSNET: The ‘WEB -SLS’ Student Journal (dedicated to PhD students)
at [URL36] where three articles have been published in six months.
Also, the Free Speech Journal (R. Cole et al., Eds [URL37]) published has
three articles since it started six months ago.
Finally, there are a few searchable databases where references to
articles may be retrieved from information about author, title content,
etc. See for instance the Frankfurt Online Bibliography in Phonetics /
Speech Technology at [URL38].
4.7
Links pages and other non-refereed articles
This category is for collections of links to relevant material, which may
be useful for researching particular topics. They often lead to descriptive
material about 'state of the art' or research activities. Since such
materials are not refereed, they can be of variable quality.
A major class of reference material found on the Web, and suitable for
student use are the "links of links" pages. These tend to be organised
around some theme and provide a large number of links to relevant web
sites. Since URL addresses change quite often, and because new pages
appear every day, these pages can get out of date, so it is important that
their maintenance is taken seriously. The content of these pages seem
to be selected according to two main criteria. One criterion is the
membership to an association or partnership of a project (e.g. see the
ESCA list of labs [URL39], although this page also gives links to non
ESCA-member laboratories.). A second criterion for selection in a list of
referred links is based on the scientific topic. General themes can be
found like "phonetics" or "linguistics" where links are often classified
according to the alphabetic order of the country (e.g. the University of
Amsterdam list at [URL40]). There are also lists of more focused topics,
like "speech recognition", "speech coding", "lipreading", "facial
animation", etc. (e.g. [URL41] or [URL42]).
4.8
Speech Science Data
This category includes acoustic, phonetic and linguistic materials
accessible over the Internet and suitable for educational purposes.
Currently the area is dominated by large corpora, and few have the
capability for downloading samples without copyright restrictions and
free of charge.
122
Computer-Aided Learning and the Use of the
Internet
Among advertised CD-ROMs (or floppies), we may highlight the UCLA
databases (the "sounds of the world” database and the UPSID database at
[URL 43]). There is also a Web presence for the big speech corpus
providers, mostly set-up for speech recognition training, see the catalogs
of the European Languge resources Association [URL44] and U.S.
Linguistic Data Consortium [URL45].
Beside these lists of existing materials available for a fee, some free
examples can be found. For historical purposes, "speech museums" can
be worth visited. Speech synthesis is ideally suited to this kind of
presentation, with stored examples or links to on-line synthesizers.
These may allow the user to download existing material or to generate
their own sentences - possibly exposing flaws in the authors’
presentation of the system. See the Museum of Speech Synthesis at
[URL 46].
4.9
Speech Science Tools
This category relates to software for speech and language analysis
suitable for educational use. There are many more references to such
tools on the Web than there are sites from where tools may be
downloaded free of charge.
A number of sound tools are freely available from the SimTel archive
[URL47] covering MS -DOS, Windows 3.1 and Windows 95 systems. These
tend to be either free or ‘Shareware’ applications, which require a
registration payment only if you find the program useful. An interesting
‘Freeware’ program is the Spectrogram program written by R.S.Horne
[URL48].
This program will display real-time spectrograms on any
reasonably fast Windows computer. However it has only primitive control
over the spectral analysis parameters, and only generates narrow -band
equivalent spectrograms for speech sampled at normal rates. The display
can be seen in Figure 5.14.
123
Computer-Aided Learning and Use of the Internet
Figure 5.14. Display from the Spectrogram program of R.S.Horne [URL48].
One of the authors of this chapter is also the main author of a suite of
speech analysis software, the Speech Filing System, available for free
download [URL49]:
“SFS provides a computing environment for conducting
research into the nature of speech. It comprises software
tools, file and data formats, subroutine libraries, graphics,
standards and special programming languages. It performs
standard operations such as acquisition, replay, display and
labelling,
spectrographic
and
formant
analysis
and
fundamental frequency estimation. It runs under Unix and
DOS environments and is currently running on Sun, HewlettPackard, Masscomp, Alpha, Linux and 486+PC.
SFS is
copyrighted University College London, but is currently
supplied free of charge to research establishments for nonprofit use. SFS is supplied as is with no warranty or support.”
Within SFS are the following tools useful for teaching purposes. Es is a
program that can display speech signals with a time-aligned wide-band
spectrogram calculated ‘on the fly’ as the user scrolls left and right and
zooms in and out. It can also print spectrograms to postscript printers.
Espect is a program to display a time waveform or wide or narrow
spectrograms along with a spectral cross section chosen by the user;
displays can also be printed to postscript printers; see Figure 5.15.
Esform is a program like Espect but which uses an LPC modelled spectral
cross section to help with formant analysis. Sprint is a program to print
spectrograms over multiple pages in a standard format of time waveform,
124
Computer-Aided Learning and the Use of the
Internet
narrow-band and wide-band spectrogram; postscript printers only are
supported.
Figure 5.15: SFS program Espect displays a wide-band spectrogram and a crosssection. It is specifically designed for use in speech science laboratories. [URL49].
5
Reflections on the Current Position
In this section we reflect on the properties of the different resources we
find in our survey. We consider their coverage, their quality, and issues
about data, tools and copyright.
5.1
Coverage
As may be expected in a new and developing field, the material in the
inventory suffers from many kinds of biases with respect to language of
instruction, geographical distribution and subject areas covered.
Language of Instruction. The overwhelming majority of the material is
in English and although there will be some bias due to our collection
procedures, the bias to English is very striking. While the dominance of
English on the Internet in general probabl y originates from the
technological domination of the U.S., there are probably other influences
on the educational resources we have discovered in Europe. These
include the use of English as the lingua franca for scientific discourse in
the field, and the wide acceptance of English as a second language in
125
Computer-Aided Learning and Use of the Internet
the Scandinavian countries.
Other languages in the inventory are
currently German, French and Swedish; however the geographical
coverage within European countries is much wider than this would
suggest.
As far as publishing on the Web is concerned, languages covered by the
‘Latin-1’ character set, are well supported in HTML. However, within
Europe, Greek and Cyrillic alphabets are not adequately supported.
There is also a particular problem with the International Phonetic
Alphabet, which must be handled as graphic images.
Geographical Distribution. Telematic courses are predominantly
although not exclusively based in the United States of America. A
similar pattern is seen for course syllabus pages. This is undoubtedly
due to the longer history of institutional networks in the U.S., and there
are plenty of signs that equivalent components are being developed in
Europe. HyperLink pages such as the World Lecture Hall based in Texas
probably have a strong U.S. bias simply due to their own geographical
position and use of English.
Other materials, tutorials, data and tools are well distributed within
Europe.
The bias to the U.K. as far as Educational Technology
information is probably due to the bias in collection; it is likely that
there are many equivalent materials in all Western-European countries.
Subject Areas. With respect to coverage of the different subject areas in
the field of Speech Communication Sciences, there is a strong bias
towards those areas where computers have been used for longest. There
are a number of tutorial materials in acoustic-phonetics and practical
phonetics, while fewer in (say) morphology or phonology. There are
materials in computational linguistics, but little in syntactic theory.
There is nothing yet in the inventory about Speech and Language
Therapy, although there are many sites for the support of the
communication disabled on the Internet.
5.2
Quality
We have not undertaken a detailed review of all the materials in the
inventory, but even a cursory inspection reveals a wide range of quality.
Some materials are clearly directed at learners, while most are
‘reference’ materials describing a particular theory or technique. Some
materials are entirely introductory, requiring little previous knowledge,
while others would only be useful if introduced by a tutor in the
student’s institution. Some are broad and shallow, others narrow and
deep.
126
Computer-Aided Learning and the Use of the
Internet
While in general this variety is to be applauded, it does raise problems
for students.
How can a student know which material is at an
appropriate level? How can a student know whether it covers the topics
he/she is required to learn? How can the student rely on the accuracy
of the description?
Again, the design of material varies with regard to its mode of
instruction: some are merely sequential descriptive, others follow
principles of ‘Open Learning’ with objectives, self-assessments and
summaries. How can the student choose the best for their own needs?
5.3
Tools
The potential author of Web-based instructional material has a mountain
to climb to become familiar with the many technological components
available on the Internet. While word-processing systems now make the
generation of basic HTML pages straightforward, there are few simple
tools to help novice authors with the interactivity and multimedia
opportunities of the Web. Teaching systems need to offer materials with
sounds, pictures and video; they should have an ability to ask questions
and mark responses; they should be able to keep details on students’
performance and status.
Current integrated tools for the building of instructional systems are
rather limited. WebCT provides an infrastructure for a telematic course,
but leaves the authoring of tutorial materials unsupported within the
framework. As a consequence, the quality of the instruction the student
receives - the heart of the educational experience - is dependent on the
technological skill of the author. This tends to limit the most exciting
materials to the most technological subjects, and makes it hard for
authors in the Arts- or Clinical- based disciplines.
A separate issue is to do with software tools for teaching Speech Science.
There is at least a 3-way split between users of Macintosh, PC/Windows
and Unix computer systems among the institutions; each with different
and incompatible program formats and audio support. The SFS software
described above requires a single sound card type on PCs since it directly
controls the hardware. Web formats for sound are multiple and can be of
indifferent quality.
While the use of authoring languages such as
HyperCard or ToolBook can make materials easier to create, they can
also add expense to the author and make the materials less portable to
different computing platforms.
We have seen at least one use of the Java language and virtual machine
in the inventory as a means to start to overcome some of these
127
Computer-Aided Learning and Use of the Internet
inadequacies. Recommendations, integrated tools and example tutorial
materials will also help in the future.
5.4
Data
Most of the references to Linguistic data in our inventory do not,
surprisingly, lead to locations where data can be downloaded!
Historically, speech data has been distributed on CD-ROM. Institutions
such as the Linguistic Data Consortium in Pennsylvania, sell many types
of data on CD-ROM and require users to sign restrictive usage licences.
The new European Language Resources Association looks as though it
will operate in a similar manner.
Thus although we might separate educational use from research use,
and particularly educational volume (a few utterances) from research
volume (hours of corpus recordings), there seems little opportunity for
students and tutors to access a suitable variety of material in multiple
languages, free of charge and free of usage restrictions.
The Linguistic Data Consortium [URL45], to their credit, do make
available some of the TIMIT phonetic corpus accessible by anonymous
FTP. However only a very poor quality version is accessible in this way.
5.5
Copyright
The issue of copyright also extends to educational material. Since the
use of the Internet for education is still quite new, the financial people
within institutions have not yet woken up the problem of the value of the
educational materials made freely available on the network. While we
don’t seem to be under threat to charge for such information at present,
it seems important that we should make a clear expression of ownership
of intellectual property on all material made globally accessible on the
network.
We need to see authors protected from exploitation by
commercial companies, while still allowing students free access.
6
Recommendations
6.1
Tutorials and Components
Our survey and inventory bear witness to the amount and range of
educational material available over the internet, largely at zero or
nominal cost. Ignoring the nature of the content for a moment, we can
order this material, in a rather crude way, according to its size and
degree of internal organisation. Tutorials and components represent
128
Computer-Aided Learning and the Use of the
Internet
particular points on a scale which stretches from simple handouts at one
extreme to complete programmes of study at the other as shown in
Figure 5.16.
Increasing size and organisation
Components
Handouts
Programme of Study
Tutorials
Figure 5.16: Size and complexity of educational resources.
Material tending towards the left end of the scale has progressively less
content and internal structure. This has advantages and disadvantages.
On the positive side is the supposed ease with which such material can
be slotted into alternative contexts, thus allowing course authors to pick
and choose as they please. For example, the categorical perception test
in section 4.4 might serve in context A to illustrate what categorical
perception is and in context B to actually create a set of categories. The
disadvantage of small units is that the size of the educational experience
they create tend to be correspondingly small. In order to create an
educational experience of any substance, a large number of such units
must be brought into play.
We choose our vocabulary carefully, since bringing into play is more than
mere juxtaposition. Bringing into play has to maintain the coherence of a
general point of view or line of argument and may involve more or less
work depending on the extent to which the individual units share
background assumptions.
Material at the other end of the scale has the opposite qualities: more
and more in the way of content, internal structure and coherence, and
less and less ability to slot into arbitrary contexts. A complete course
provides the self-contained educational experience but you are more or
less forced to take it or leave it as it is, since generally the cost of
adaptation is higher than the cost of rewriting from scratch.
The question is whether there is a point on the scale which maximises
the advantages and minimises the disadvantages. We believe that
tutorials, being positioned somewhere around the middle of the scale, lie
at just such a point.
129
Computer-Aided Learning and Use of the Internet
6.2
Key Features Of Tutorials
A tutorial typically offers a structured learning experience of limited
dimensions. In many ways a tutorial is just like a course with certain
restrictions and general characteristics:
• Or i ent ed t o Sel f-Learning : involves only the student and the material.
Direct interaction with third parties is usually unnecessary to follow
the tutorial successfully.
• Asynchronous:
interaction
with
the
tutorial
takes
place
intermittently; moreover, the student decides when interaction will
take place, not the tutorial. Hence tutorials are also self paced.
• Rest r i ct ed T opi c and Pr er equi si t es: A tutorial concerns a clearly
defined topic or group of related topics and has limited aims. The
prerequisites associated with tutorials are usually minimal, but this
should not exclude the idea of prerequisites altogether. What is
important is that the tutorial author has a clear idea of what
background the students are expected to have, and conversely that the
student knows what background is expected.
• Restricted Conceptual Difficulty: Essentially, a tutorial is not an
advanced course. This could be manifested in many ways, e.g. though
the use of explanatory material, the use of examples, the avoidance of
presumptuous forms which assume a great deal of background
knowledge. Of course, it is still possible to have tutorial level material
in a difficult subject (this is precisely its raison d’être).
• Defi ned Int er nal St r uct ur e: although some tutorials comprise a
single chunk of text or hypertext (e.g. the Speech Analysis Tutorial
[URL50]), the tutorial concept is big enough to accommodate more
sophisticated kinds of structure (e.g. the "chapitres" of the Lausanne
Phonetics course [URL51])).
• Completeness: by and large, tutorials are self contained. However,
being of a certain size, they are large enough to embody not just a
piece of subject matter, but a ``perspective" that might extend through
several tutorials. When this is the case, we can envisage larger units
of study that are created by composing several tutorials together. A
very good example of this is provided by the Java Tutorial [URL52]
which is organised into "trails": groups of lessons on a particular
subject which tell a different kind of story about the same topic.
• Self Assessment : An important aspect of self-learning is selfassessment, the ability to judge one's mastery of the material being
presented.
Many tutorials make a point of providing for selfassessment in the form of exercises. The advent of languages like
130
Computer-Aided Learning and the Use of the
Internet
Java and JavaScript considerably enhance the potential for interactive
assessment exercises to take place over the internet.
6.3
Description and Indexing of Tutorials
One of the aims of the thematic network is to pr omote a common
educational framework for the Speech Communication Sciences. The
indexing of not just tutorials, but all types of educational materials, can
only make sense in the context of such a framework. n integrated
curriculum, which, from a functional perspective, can be usefully
regarded as a kind of map with two projections, each of which serves a
different purpose.
One of them, the "conceptual" projection, concerns the field of study,
indicating how different parts of the subject as a whole relate to each
other. The other projection (we could call this the "programme"
projection), is essentially composed of pedagogical elements such as
courses, tutorials, demonstrations: in short, items fitting on the scale
shown in Figure 5.16.
The projection gives information about the
relations between these elements. Examples of such relations might be:
item X is a prerequisite for item Y; item X can be used to assess
performance of property P in item Y. From the information in this
projection we can derive a set of possible programmes (only some of
which will be realisable).
Many of the problems of indexing course materials will go away, or at
least be alleviated, if a common framework exists. Procedures should
therefore be instigated to make it exist sooner rather than later.
Whether or not the above scheme is adopted in the exact form just
outlined, it is important to stress that in general, formal or semi-formal
representations are less likely to be misunderstood than informal ones,
and are therefore more likely to serve as the basis for a common
language.
6.4
Providing Encouragement
Our survey has shown us a great deal of what is possible and what is the
potential for computer-based methods.
Future activities should be
directed towards ensuring that potential is realised. This can be done in
a number of ways:
• Help students and tutors find material. We should enhance our
existing repository and make it more widely known.
• Design a set of tutorial ‘characteristics’ so that users can find
tutorials of an appropriate content and depth.
131
Computer-Aided Learning and Use of the Internet
• Encourage the creation of quality tutorials by publishing guidelines
based on Open Learning principles (Race, 1994)
• Reward the authors of good tutorial materials with recommendations
Acknowledgements
We are grateful for the help of other members of the Working group and
other members of the Thematic Network in the preparation of this
Chapter. We thank all those that contributed information about educational resources through our questionnaire.
132
Computer-Aided Learning and the Use of the
Internet
Uniform Resource Locators
1. http://www.york.ac.uk/inst/ctipsych/web/PsyCLE/PsyCLE.html
2. http://www.realaudio.com
3. http://www.icbl.hw.ac.uk/tltp
4. http://www.cti.ac.uk
5. http://www.icbl.hw.ac.uk/ltdi
6. http://www.icbl.hw.ac.uk/tltsn
7. http://www.icbl.hw.ac.uk/ltdi/man-info/umi.html
8. http://www.talisman.hw.ac.uk/
9. http://www.phon.ucl.ac.uk/home/netphon/Welcome.html
10. http://ukonln.bath.ac.uk/elib/
11. http://www.netskills.ac.uk/
12. http://weber.u.washington.edu/~dillon/adamn1.html
13. http://www.cse.ogi/CSLU/cse551/guessspec.html
14. http://svr-www.eng.cam.ac.uk/ajr/SpeechAnalysis/
15. http://www.research.att.com/
16. http://medicus.marshall.edu/medicus.htm
17. http://cvu.strath.ac.uk/admin/what-is-CVU .html
18. http://www.uhi.ac.uk/
19. http://tn-speech.essex.ac.uk/tn-speech/project/groups/tn-cal/quest.html
20. http://tn-speech.essex.ac.uk/tn-speech/project/groups/tn-cal/tn-cal.html
21. http://www.csuhayward.edu/ics/htmls/Inst.html
22. http://www.edb.utexas.edu/coe/depts/ci/it/projects/papers/special.html
23. http://homebrew.cs.ubc.ca/webct/
24. http://www.phonetik.uni-muenchen.de/AP/APHome.html
25. http://www.music.mcgill.ca/auditory/Auditory.html
26. http://ophale.icp.grenet.fr/
27. http://www.phonetik.uni-muenchen.de/Haskins/Haskins/MISC/PP/pp.html
28.
http://www.phonetik.uni-
muenchen.de/Haskins/Haskins/MISC/ASY/asy.html
29. http://www.ling.umu.se:80/~anderse/education/Katper.html
30. http://www.utexas.edu/world/lecture/linguistics/
31. http://www.bucknell.edu/departments/linguistics/ln105.html
32. http://www.cse.ogi.edu/CSLU/HLTsurvey/HLTsurvey.html
33. http://www.ilc.pi.cnr.it/EAGLES96/browse.html
34. http://ophale.icp.grenet.fr/esca/ProgAVSP.html
35. http://www.elsevier.nl:80/cas/tree/store/specom/free/noncas/
audio/menu.sht
36. http://web-sls.essex.ac.uk/web-sls/
37. http://www.cse.ogi.edu/CSLU/fsj/home.html
38. http://www.uni-frankfurt.de/~ifb/bib_engl.html.
39. http://ophale.icp.grenet.fr/esca/labos.html
40. http://fonsg3.let.uva.nl/Other_pages.html
41. http://svr-www.eng.cam.ac.uk/comp.speech/SpeechLinks.html
42. http://mambo.ucsc.edu/
43. http://www.humnet.ucla.edu/humnet/linguistics/faciliti/software.htm
44. http://www.icp.grenet.fr/ELRA/home.html
133
Computer-Aided Learning and Use of the Internet
45. http://www.ldc.upenn.edu/
46. http://ophale.icp.grenet.fr/ex.html
47. http://sunsite.doc.ic.ac.uk/Mirrors/simtel.coast.net/SimTel/win3/sound/
48. ftp://sunsite.doc.ic.ac.uk/computing.systems/ibmpc/windows95/
sound/gram32.zip
49. http://www.phon.ucl.ac.uk/resource/sfs.htm
50. http://www.ling.lu.se/research/speechtutorial/tutorial.html
51. http://www.unil.ch/ling/phonetique/api.html
52. http://java.sun.com:80/docs/books/tutorial/
134
Computer-Aided Learning and the Use of the
Internet
Appendix I - CAL/Internet Resources Inventory
The working group has set up a Web-based database of teaching
resources, to which individuals can contribute through a forms interface
accessible by Web browser. Each night, the contents of the database are
converted to HTML so that it may be viewed from anywhere on the
Internet. The database currently contains details of over 80 resources,
and in this chapter we have categorised the types of material it contains.
We made the decision to focus on resources explicitly designed for
educational purposes, and on materials that were accessible directly
over the Internet. In future years we may widen our review to reference
materials and to commercial materials. The following listing summarises
the contents of the database.
1
General Information Category
Delivering Instruction on the World Wide Web
Paper on basic issues involved in Web based instruction, design and
delivery.
URL:http://www.csuhayward.edu/ics/htmls/Inst.html
Using the WEB as an Instructional Tool
Overview of Web use for instruction in the University of Oregon.
Discusses general issues about Web-based delivery.
URL:http://zebu.uoregon.edu/webreport.html
Special Considerations for Designing Internet Based Instruction
This paper addresses the questions: when is the Internet the appropriate
technology for your instruction, and what do designers need to know to
use it?
URL:http://www.edb.utexas.edu/coe/depts/ci/it/projects/papers/specia
l.html
The Art of Developing On-line Courses
Guidelines for internet delivery based on sound instructional
development principles. Contains links to other related material.
URL:http://www.cpcc.cc.nc.us/on_line/develop.htm
Teaching with the Web
Compilation of ideas for using WWW resources as a teaching tool .
URL:http://polyglot.lss.wisc.edu/lss/lang/teach.html
Interactive Courseware Tutorial
Over the past thirty years, computers have aided in the delivery and
management of instructional material. The process has been evolving
135
Computer-Aided Learning and Use of the Internet
with an emphasis on building a more effective, interactive instructional
system. The early use of computers as an instructional delivery media
was characterised by limited student control of instructional sequence
and little interaction with the instructional content. Programmed text
was presented to the student with prescribed drills/practice and preprogrammed answers and remedial comments. With technological
advances the level of interaction and student control has increased. With
advances in technology, the emphasis is on the development of
intelligent instructional and tutorial models.
URL:http://ott.sc.ist.ucf.edu/1_2/index.htm
Learning Technology Dissemination Initiative
The Learning Technology Dissemination Initiative is a project funded by
the Scottish Higher Education Funding Council to promote the use of
learning technology and computer based learning materials.
URL:http://www.icbl.hw.ac.uk/ltdi/
Teaching and Learning Technology Programme
These pages provide information about the Teaching and Learning
Technology Programme (TLTP) along with extracts from Newsletters and
information releases. They act as a central access point for information
servers run by other projects within TLTP and related areas.
URL:http://www.icbl.hw.ac.uk/tltp/
Web-Based Training Information Centre
The W eb-Based Training Information Center is a non-profit resource for
individuals and organisations interested in developing and delivering
training using Web technology. The intent of this site is to share nonproprietary information, stimulate creative ideas, and link to interesting
training sites around the world.
URL:http://www.clark.net/pub/nractive/wbt.html
CALISCENET
Caliscenet : a network of correspondents sharing interest in computer
aided learning and instruction in science and engineering.
URL:http://diwww.epfl.ch/w3leao/calnet1.html
TONIC: Online Netskills Interactive Course
The course provides an introduction to the Internet and computer
networks in general, describing and illustrating the main software tools
for navigating the networks. The course looks at types and examples of
networked information, at the means for searching that information, and
at the communication facilities and resources on the net.
URL:http://www.netskills.ac.uk/TONIC/
136
Computer-Aided Learning and the Use of the
Internet
Open University Institute of Educational Technology
The Institute is the largest centre for educational technology in the
world and has top rankings for its contributions to teaching and for its
research. In addition to its work on courses across the Open University,
it is engaged in many collaborative projects and consultancies, in all
parts of the world, as part of its mission and objectives.
URL:http://www-iet.open.ac.uk/iet/iet.html
WWW: Background Information
WWW: BACKGROUND INFORMATION The purpose of this document is to
provide background information on WWW (World-Wide Web), HTTP
(Hypertext Transfer Protocol), and HTML (Hypertext Mark-up Language).
Information regarding NCSA Mosaic and other WWW browsers, plus
material on other important aspects of cyberspace such as gopher and
WAIS is covered.
URL:http://www.ncsa.uiuc.edu/SDG/Software/Mosaic/Docs/wwwinfo.html
2
General Tools
WebCT World Wide Web Course Tools
WebCT is a tool that facilitates the creation of sophisticated WWW based
educational environments.
URL:http://homebrew.cs.ubc.ca/webct/
Multimedia Authoring Information
Covers: 1. Packages - sites with information on the main packages.
2. Prices - how much are these packages going to cost you? 3. Which
authoring package? - an on-line argument on their faults and merits.
4. Which platform? - should you be developing multimedia on Mac or PC?
5. Discussion lists - further help available on designing multimedia.
URL:http://www.strath.ac.uk/CAL/author.html
Shockwave download Center / MACROMEDIA
Shockwave is interactive multimedia on the Web. It allows animated
interfaces, interactive presentations, streaming CD-quality audio.
Shockwave player delivers animation, audio, and multimedia. The
streaming Shockwave player is available for free from Macromedia.
URL:http://www.macromedia.com/shockwave/
Authorware by MACROMEDIA
Authorware is an authoring environment for creating and publishing
interactive information. Authorware provides tools to create networked
interactive multimedia training and educational courseware.
URL:http://www.macromedia.com/software/authorware/
137
Computer-Aided Learning and Use of the Internet
W3 Search Engines
This documents collects some of the most useful search engines
available on the WWW.
URL:http://cuiwww.unige.ch/meta-index.html
3
Tutorials
Akustische Phonetik
Course material on acoustic phonetics
URL:http://www.phonetik.uni-muenchen.de/AP/APHome.html
En tur i fonetikens marker
This is a tutorial on linguistic and paralinguistic phenomena in speech. It
touches the following topics: F0; formants; Swedish vowels; linguistic,
expressive, organic, and transmittal factors shaping speech signals; the
importance of F0 for the perception of these different kinds of qualities,
especially in vowels; manipulations of paralinguistic quality; the
modulation theory of speech; the role of expectations and perception by
compatibility testing.
URL:http://www.ling.su.se/staff/hartmut/tur.htm
Web-based course in Corpus Linguistics
This is a free web-based course in Corpus Linguistics designed to provide
a brief introduction to the area. It is based on a cut-down version of the
book on Corpus Linguistics published by Edinburgh University Press. The
course was developed as part of a collaborative venture between
Lancaster University and Edinburgh University Press under the
Innovation in Higher Education programme.
URL:http:/www.ling.lancs.ac.uk/monkey/ihe/linguistics/contents.htm
Phonetics and Phonology Tutorial
Covers: 1. Phonetics: Overview of Phonetics, Consonants, Vowels,
Phonetic Examples, Phonetic Transcription/American English. 2.
Phonology: Overview of Phonology, Phonological Rules, Phonological Rules
Practice, Autosegmental Phonology.
URL:http://www.csulb.edu/~phoneme/pronunciation.html
Speech Analysis Tutorial
A brief introduction to waveforms, spectrograms and other speech
analyses.
URL:http://www.ling.lu.se/research/speechtutorial/tutorial.html
A Course in Phonetics
These disks include nearly all the material in the tables and exercises
in the textbook A Course in Phonetics. 16 DSDD disks, requiring about 25
meg of disk space when expanded.
138
Computer-Aided Learning and the Use of the
Internet
URL:http://www.humnet.ucla.edu/humnet/linguistics/faciliti/software.ht
m
Acoustics Phonetics Teaching Materials / UCLA
This package includes two versions of the stack ‘Acoustics for
Phoneticians’, one for students in which there is a lot of explanation on
each card, and one for instructors to use in class, with fewer words in
large type.
URL:http://www.humnet.ucla.edu/humnet/linguistics/faciliti/software.ht
m
Cours de phonétique de la section de linguistique de l'Université de
Lausanne
Le but de ce document est de servir de support à l'apprentissage de la
phonétique et à la maîtrise de l'Alphabet Phonétique International
(A.P.I.). Vous pouvez parcourir cet (hyper)texte à votre guise, soit
linéairement du premier chapitre au dernier, en sautant directement au
chapitre qui vous intéresse grâce à la table des matière.
URL:http://www.unil.ch/ling/phonetique/api.html
On-line phonology course
The course is designed so that either native or non-native English
speakers can achieve quite a sophisticated understanding of the basics of
phonology. It is designed as a teaching aid rather than as a definitive
study of phonology and, as such, sacrifices detail in favour of clarity and
ease of understanding.
URL:http://www.stir.ac.uk/epd/celt/staff/higdox/stephen/phono/phono
lg.htm
Einführung in die Sprachsynthese
Introductory text on speech synthesis with a link to the Haskins Pattern
playback page.
URL:http://www.phonetik.uni-muenchen.de/HS/Synthese.html
CSLU Spectrogram Reading Home Page
Covers: Why would anyone want to read a spectrogram? What are
waveforms? What are spectrograms? What are phonemes? Improve your
spectrogram reading skills. Spectral cues for English phonemes. View
spectrograms for common words. Call the CSLU and see your own voice.
The current mystery spectrogram.
URL:http://www.cse.ogi.edu/CSLU/cse551/
Das Lesen von Sonagrammen V0.2
Course material on spectrogram reading
URL:http://www.phonetik.uni-muenchen.de/SGL/SGLHome.html
139
Computer-Aided Learning and Use of the Internet
Automatische Spracherkennung - Grundlagen, statistische Modelle und
effiziente Algorithmenn
Lecture notes for a course on speech understandíng based on the book
"Automatische Spracherkennung - Grundlagen, statistische Modelle und
effiziente Algorithmen" by Ernst Günter Schukat-Talamazzini.
URL:http://www5.informatik.unierlangen.de/HTML/English/Lectures/sprachverst/sprachverst.html
Language and Brain / Sprache und Gehirn
An audio-visual tutorial on clinical linguistics and phonetics. Contents:
Introduction to the motor system and neuro imaging, description of
language and speech pathologies, language and speech diagnostics,
glossary and extensive bibliography. The tutorial is mostly in German.
URL:http://www.ims.uni-stuttgart.de/phonetik/joerg/sgtutorial/
4
Component Resources
Categorical Perception
A perception test, replicating a classical experiment on Categorical
Perception.
URL:http://www.ling.umu.se:80/~anderse/education/Katper.html
F0 and vowel perception
This is a demonstration of the importance of the frequency of the voice
fundamental for the perception of the linguistic, expressive, and organic
qualities of vowels. It is shown that not only the organic and expressive
qualities of a vowel, but also its linguistic quality varies widely with F0.
URL:http://www.ling.su.se/staff/hartmut/i.htm
Equations for CB-rate, CB, and ERB
The most accurate equations for Hz to bark conversion and inverse
presented together with some reasoning about when it is correct to use
CB-rate and ERB.
URL:http://www.ling.su.se/staff/hartmut/bark.htm
3D modelling of visible speech gestures
Presentation of several 3D models of facial components controlled by
speech-oriented command parameters.
URL:http://ophale.icp.grenet.fr/2.6.html
Description of ToBI Prosodic Labelling System
ToBI (for Tones and Break Indices) is a system for transcribing the
intonation patterns and other aspects of the prosody of English
utterances.
URL:http://ling.ohio-state.edu/Phonetics/E_ToBI/etobi_homepage.html
140
Computer-Aided Learning and the Use of the
Internet
The Pattern Playback
A description (w. sound clips in .au and -aiff) of the Pattern Playback
`talking machine', an early speech synthesiser constructed in the late
1940s.
URL:http://www.phonetik.unimuenchen.de/Haskins/Haskins/MISC/PP/pp.html
UCSC Perceptual Science Laboratory
Information is available on the following topics: Facial Animation, Facial
Analysis, Speechreading (Lipreading).
URL:http://mambo.ucsc.edu/
Click and Listen
"Click and Listen" is a network-based teaching resource for phonetics
and phonology, with particular emphasis on the sounds of Scots and
Scottish English. The WWW site contains examples of the type of
material being developed.
URL:http://wheecher.arts.ed.ac.uk/
5
Course Information
Linguistics 105 Sounds and Words
This is an on-line interactive syllabus for a course in linguistic
phonetics.
URL:http://www.bucknell.edu/departments/linguistics/ln105.html
World Lecture Hall/Linguistics
Hyperlinks to syllabus information and telematic courses in Linguistics,
Phonetics and Spoken Language Processing.
URL:http://www.utexas.edu/world/lecture/linguistics/
Ling 253: Laboratory Phonetics
Interactive Course Syllabus: Information, Lecture Notes, Resources,
Speech Files, Laboratory
URL:http://gopher.udel.edu/idsardi/253.htm
Acoustic Phonetics Slides
Slide show of a course in Acoustic Phonetics
URL:http://eras-speech.essex.ac.uk/erasspeech/syllabus/acoustic.html
Language, Mind and Society Course
Syllabus and Course Notes, with Articulatory Phonetics tutorial.
URL:http://macweb.acs.usm.maine.edu/usm_linguistics/Course%20Mat
erials/185Language%2cMindandSociety/LIN185_F96/185_F96.html
141
Computer-Aided Learning and Use of the Internet
6
Bibliographies and Refereed Articles
Online Bibliography: Phonetics and Speech Technology
Search for books and articles on subjects allied to Phonetics and Speech
Technology in this bibliography with over 10,000 entries.
URL:http://www.uni-frankfurt.de/~ifb/bib_engl.html
Content of the Speech Communication Journal
Gives a list of the articles published in the "Speech Communication"
Journal since 1994.
URL:http://www.elsevier.nl/eee/specom/contents.html
WEB-SLS The European Student Journal of Language and Speech
Web SLS is a joint initiative of ESCA, EACL & ELSNET Given the wide
coverage of the Journal in terms of scientific areas, nine very broad
classes on which to concentrate have been selected: SRS - speech
recognition and synthesis NLUG - natural language understanding and
generation LI - language identification IER - information extraction and
retrieval DMC - dialogue modelling and control MLPP - models of
language perception and production SM - stochastic modelling GCLD grammars, corpora, lexicons and databases SAS - system assessment
strategies
URL:http://web-sls.essex.ac.uk/web-sls/
The Free Speech Journal
The World Wide Web provides an excellent opportunity to fix a serious
problem in science -- the long delay in publishing research results in
scientific journals. Typically, more than a year goes by, often two,
between the time a manuscript is submitted for publication and the time
it appears in print. The Web provides an efficient means of
communicating research advances and new ideas, and an opportunity to
reduce publication delay by an order of magnitude. The motivation for
founding the Free Speech Journal (FSJ) is to improve communication
among scientists in the area of human language technology. This is done
by providing a peer review journal (FSJ) for publishing research results of
the highest quality in the shortest possible amount of time.
URL:http://www.cse.ogi.edu/CSLU/fsj/home.html
Computation & language EPrint Archive
Archive of academic papers from researchers
computational linguistics.
URL:http://xxx.lanl.gov:80/cmp-lg/
142
in
NLP,
NLE
and
Computer-Aided Learning and the Use of the
Internet
7
Links pages and non-refereed articles
Speech Technology Hyperlinks
Probably the biggest list of speech technology links available. From the
comp.speech newsgroup Frequently Asked Questions list.
URL:http://svr-www.eng.cam.ac.uk/comp.speech/SpeechLinks.html
Linguistics & Phonetics Worldwide
Extensive structured list of language and speech related hyperlinks.
URL:http://www.ims.unistuttgart.de/phonetik/joerg/worldwide/lingphon.html
List of European & worldwide Speech Laboratories
List of home pages referring to Research Laboratories in Phonetics &
Speech Communication. This list is produced by ESCA, the European
Speech Communication Association.
URL:http://ophale.icp.grenet.fr/esca/labos.html
Speech on the Web
Links to many pages that are related to phonetics and speech sciences:
Meetings, Lists, Phonetics, Language Processing, Computational
Linguistics, Dictionaries, Newsletters
URL:http://fonsg3.let.uva.nl/Other_pages.html
The Human Languages Page
Information about the languages of the world. Language resources,
dictionaries, tutorials, samples.
URL:http://www.dcs.warwick.ac.uk/~bear/Language-Page.html
Ethnolinguistics and Phonetics on the Internet
An ordered list (on keyword and title) of Internet sources with United
Resource Locator, summary and additional comments.
URL:http://grid.let.rug.nl/~degraaf/EthnoPhon/linklist.html
Links to Hypermedia demonstrations and collections
Links to Hypermedia demonstrations and collections: Speech Synthesis,
Speech Recognition, and others
URL:http://fonsg3.let.uva.nl/IFA-Features.html
European Language Resources Association
ELRA is a non-profit association who disseminates speech & language
resources
URL:http://www.icp.grenet.fr/ELRA/home.html
The International Phonetic Association
Activities of the International Phonetic Association
URL:http://www.arts.gla.ac.uk/IPA/ipa.html
143
Computer-Aided Learning and Use of the Internet
Computer Assisted Pronunciation Investigation Teaching and Learning SIG
CAPITAL is a group of researchers and practitioners interested in using
computers in the domain of pronunciation.
URL:http://showme.missouri.edu/~langdans/capital.html
Resources for Studying Human Speech
The Internet is rich in resources for the study of speech sounds. Here
are some that I have collected and used. You should be able to use some
of them at least online if your system supports the replay of AU or WAV
sound files.
URL:http://weber.u.washington.edu/~dillon/PhonResources.html
8
Speech Science Data
British National Corpus of English
The British National Corpus is a very large (over 100 million words)
corpus of modern English, both written and spoken.
URL:http://info.ox.ac.uk/bnc/
MARSEC: Machine Readable Spoken English Corpus
MARSEC is a CD-ROM of the recordings transcribed in the Spoken
English Corpus (SEC). Annotations at word, grammatical and prosodic
levels are available.
URL:http://midwich.reading.ac.uk/research/speechlab/marsec/marsec.
html
Oxford Acoustic Phonetic Database
Available on compact disc, from J. Pickering and B. Rosner. It contains
data on vowel-consonant and consonant-vowel combinations in both
stressed and unstressed locations. The language covered include
French, German, Hungarian, Italian, Japanese, British English, Spanish
and English.
URL:http://svrwww.eng.cam.ac.uk/comp.speech/Section1/Data/oxford.html
The Sounds of the World's Languages
15 DSDD disks, requiring about 35 meg of disk space when expanded
Sounds of the World's Languages (SOWL) is a Hype rcard tool and
database developed at the UCLA Phonetics Laboratory to illustrate and
teach about the range of sounds used in human languages with material
on more than 80 languages.
URL:http://www.humnet.ucla.edu/humnet/linguistics/faciliti/software.ht
m
144
Computer-Aided Learning and the Use of the
Internet
UPSID and PHONEME
The UCLA Phonological Segment Inventory Database of (currently) 451
languages, with programs to manipulate and modify it.
URL:http://www.humnet.ucla.edu/humnet/linguistics/faciliti/software.ht
m
Relator project speech corpora index
This is an index page describing a large number of speech databases
through Europe.
URL:http://www.icp.grenet.fr/Relator/countries.html
EUR-ACCOR acoustic and articulatory database
ACCOR is a unique acoustic and articulatory database recorded as part
of the ESPRIT- ACCOR project investigating cross-language acousticarticulatory correlations
in
coarticulatory processes.
European
Languages include: Catalan, English, French, German, Irish Gaelic,
Italian and Swedish.
URL:http://www.icp.grenet.fr/Relator/multiling/euraccor.html
Audio files associated with articles in the Speech Communication Journal
Audio files associated with articles published in the journal "Speech
Communication"
URL:http://www.elsevier.nl:80/cas/tree/store/specom/free/noncas/au
dio/menu.sht
Linguistic Data Consortium Speech Corpora
The Linguistic Data Consortium is an open consortium of universities,
companies and government research laboratories. It creates, collects and
distributes speech and text databases, lexicons, and other resources for
research and development purposes. The LDC is currently collecting
thousands of recorded telephone conversations among speakers of more
than ten different major languages of the world. Calls are initiated from
the United States or Canada and made to all parts of the globe.
URL:http://www.ldc.upenn.edu/
Speechlab / Sprachlabor
Speechlab is a multimedia tutorial on phonetics, containing chapters on
articulation, acoustics, speech analysis, and a multimedia reference of
American and German speech sounds (video, audio, articulatory and
acoustic description). Also included are a small speech analysis tool and
a bibliographic database containing over 4000 articles. Sprachlabor is the
German Version, without the American sound lexicon.
URL:http://www.media-enterprise.de
WordNet a lexical database for English
On-line and downloadable version of a large lexicon.
URL:http://www.cogsci.princeton.edu/~wn/
145
Computer-Aided Learning and Use of the Internet
9
Speech Science Tools
Speech Filing System
The Speech Filing System (SFS) is a large collection of inter-working
tools for speech analysis, processing and display. It used both for
research and in our teaching laboratory at University College London.
URL:http://www.phon.ucl.ac.uk/resource/sfs.htm
Czech TTS System for Windows
Czech Text-To-Speech system for Windows. As user application can read
files, selected text in any application (via clipboard) and as MCI driver.
Application interface: DDE communication. Text to Speech Engine
(W95,NT only) form in progress.
URL:http://www.anet.cz/frog
WinSAL and WinSAL-V
WinSAL is a speech analysis system for Windows 95. It allows to load,
segment and display multiple audio files in the wav format. Functions
include oscillogram, spectrogram, Energy, Pitch, FFT, LPC, Cepstrum,
Autocorrelation, AMDF. Data can be saved to text files, screen
hardcopies are possible. WinSAL-V additionally allows to analyze video
files in the AVI format.
URL:http://www.media-enterprise.de
Natural language software registry
Software for speech signal analysis, syntactic analysis, formalisms,
generation, knowledge representation, applications.
URL:http://cl-www.dfki.uni-sb.de/cl/registry/draft.html
Intelligent Speech Analyser
The intelligent speech analyser is a software tool running on Macintosh
computers. It allows a large number of types of display of speech signals
including waveform, FFT, cepstrum, LPC, Auditory spectrum, long-term
average spectrum, F0, Jitter, Shimmer, etc. It is unclear as to the cost
from the home pa ge.
URL:http://www.sci.fi/~pitchsys
SAM Speech Tools
This is a list of speech tools produced during the ESPRIT 2589 "SAM"
Project. It consists mainly of software tools running on a standard SESAM
workstation (PC-DOS based) Labelling, Data collection, Signal editing,
Processing, DBMS, Testing - Scoring , Recognition.
URL:http://www.icp.grenet.fr/Relator/tools/spbacktools.html
146
Computer-Aided Learning and the Use of the
Internet
SIL Software for Linguistics
Summer Institute Of Linguistics: Computing Resources, including:
publications, bibliographies, research projects, training, tools, fonts.
URL:http://www.sil.org/computing/sil_computing.html
Lucent Technologies Bell Labs Text-to-Speech system
The site has interactive demos in American English, Mandarin Chinese,
and German, and the capability to adjust voice parameters on the fly
(English only). Pre-synthesized demos for Mandarin Chinese, French,
Italian, Russian, Spanish, and Romanian are also provided.
URL:http://www.bell-labs.com/project/tts/
Articulograph AG100 Demo - Programs for downloading
Emalyse : Visualisation - Analysis - Segmentation Tailor : Editing - Postprocessing - Normalisation MultiCV: Format conversion - data export. The
Emalyse is well equipped with features to show all information from the
recorded data. It makes it easy and secure to places markers and
segments with high precision at the desired positions. The modifications
on the data are only done by the Tailor program. It does the general data
post-processing and is also able to read the Emalyse marker- and
segment information to do the required data manipulation. The MultiCV
changes the format from the data but does not change the data itself. It
makes the data readable for general data base and calculation programs
as well as graphic presentation programs.
URL:http://www.articulograph.de/demo_p.htm
IPKoeln: Speech synthesis
An articulatory speech synthesizer has been developed at our phonetics
department. The input of the program is a phonetic symbol string and its
output is an acoustic speech signal. The program can be divided into
three subsystems. The first subsystem calculates a movement pattern for
the model articulators (lips, tongue, velum, glottis) on the basis of the
symbol string (broad phonetic transcription). For this calculation a
segmental or a gestural production model can be used. The second
subsystem is the articulator model. It calculates the geometry of the
vocal tract (the area function) for each time instant. The third subsystem
is the aerodynamic-acoustic model. It calculates the acoustic speech
signal.
URL: http://www.rrz.uni-koeln.de/philfak/phonetik/synthese/index_e.html
COMPOST
COMPOST is a complete environment for text-to-speech developers. It
offers: - A rule-compiler where declarative (grammars) and procedural
(external calls to C code) knowledge can be combined. The standard
library offers a Morphological Analysis, a N-class Syntactic filter, a TDPSOLA synthesizer... - COMPOST is built as a client-server model: a
147
Computer-Aided Learning and Use of the Internet
compiled scenario for any language may be loaded by the server and used
by clients to convert text into speech - Scenarios have been developed for
French, Catalan and Castillan
URL:http://www.icp.grenet.fr/cgi-bin/synthese
Text-to-audiovisual-speech
Presentation of a text-to-audiovisual-speech synthesizer whose front-end
is a parametric model of the face.
URL:http://ophale.icp.grenet.fr/2.1.html
148