JP2005182127A - Information processor and information processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To preferentially output a sentence in which a user is highly interested concerning contents including a plurality of sentences. <P>SOLUTION: While a contents outputting part 8 outputs contents, a user request inputting part 4 inputs a request regarding a contents viewing operation from a user (that is, a request for the repetition of the output or the skip of the output of a specific sentence), and a state deciding part 9 decides the sentence being the target of the contents viewing operation request, and a user information generating part 10 counts points for words included in the sentence being the target of the contents viewing operation request, and generates user information representing the points of each word, and a contents analyzing part 6 compares the points of each word represented by the user information concerning each sentence of the contents received by a contents receiving part 5 with words included in each sentence, and decides the output sequence of the sentence in order to preferentially output the sentence whose point is high and a contents output part 8 outputs the sentence in the decided output order. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a Web browser that can be operated by voice operation on a communication terminal.

  WWW (World Wide Web) browsers that can acquire various contents existing on the Internet, display them visually, and provide various information by providing links to the following information are generally widespread Yes. In addition, WWW browsers that can reproduce and operate content by voice recognition, voice synthesis, etc. have appeared, regardless of visual display or input devices such as a mouse and a keyboard.

For example, “Browser system, voice proxy server, link item reading method and storage medium storing link item reading program” disclosed in Japanese Patent Application Laid-Open No. 11-110186 discloses a text included in WWW information as a target of voice operation. Can be read out in order.
JP-A-11-110186

  In the prior art, even if the content can be played back by voice or the like, depending on the content content, if the information is not intended by the user, or if the information required by the user is described later in the content, I had to wait for it to be output.

  In addition, the information intended by the user cannot be selected from various information.

  Accordingly, an object of the present invention is to solve such a problem.

An information processing apparatus according to the present invention includes:
A content receiver that receives content including at least one sentence;
A content output unit that outputs sentences included in the content according to a predetermined output order;
A user request input unit for inputting a content viewing operation request from a user who uses the content;
An output order reference generation unit that generates an output order reference used to determine the output order of sentences based on a history of content viewing operation requests input by the user request input unit;
And an output order determining unit that determines an output order of each sentence included in the content received by the content receiving unit using the output order reference generated by the output order reference generating unit.

  According to the present invention, the output order reference is generated based on the history of the content viewing operation request, and the output order of each sentence included in the content is determined using the generated output order reference. Information with high interest can be preferentially output.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a basic configuration of browser apparatus 1 (information processing apparatus) according to the first embodiment.

  In the figure, a content receiver 5 receives content from the Internet or the like by communication.

  The content analysis unit 6 (output order determination unit) analyzes the received content. The analysis is a process of scanning the content, determining the document structure and content breaks for display, and outputting the analysis data. FIG. 2 shows an example of analysis. Headline information and content information are extracted by analyzing HTML content described in HTML or the like. When the content description example shown in FIG. 2 is displayed, it is displayed as shown in the display example. Further, each of the contents information of contents 1 and 2 shown in FIG. 2 corresponds to a sentence. For this reason, the browser device 1 according to the present embodiment receives content including a plurality of sentences. As will be described later, the content analysis unit 6 determines the output order of each sentence so as to output from the sentence in which the user of the browser device 1 is interested. The content after the output order is determined by the content analysis unit 6 is referred to as analysis information.

  The analysis information storage unit 11 temporarily stores the content after the output order is determined by the content analysis unit 6, that is, analysis information, and outputs it to the content output unit 8.

  The content output unit 8 is a part that inputs analysis information from the analysis information storage unit 11 and outputs the analysis information so that the user can visually see it or hear it as sound. In addition, when outputting audio information, it is possible to repeatedly output a specific part of content in response to an operation request from the user, or to cancel (skip) output of a specific part of content. .

  The user request input unit 4 inputs general Web browsing operation requests such as “return” and “forward”, jump operation to the next information, and input operation to the form. In addition to operations on the screen using a mouse / keyboard or the like, “next” or “previous” may be input using voice recognition processing. Further, the user request input unit 4 performs an output repeat request for requesting repeated output of a specific part of the content or an output cancel requesting output cancellation (skip) of the specific part of the content by a user's mouse / keyboard operation or voice recognition processing. A request can be input from the user. An operation request related to content viewing input from the user by the user request input unit 4 is referred to as a content viewing operation request. In the following description, an output repeat request and an output cancel request will be described as examples of content viewing operation requests. However, requests for “back”, “forward”, and jump operations to the next information can also be handled as content viewing operation requests. it can.

  The state determination unit 9 determines the content output state when a content viewing operation request is input. When audio output is being performed, the status such as which part of the content is currently being output when the “repeat” operation is designated.

  The user information generation unit 10 (output order reference generation unit) determines what kind of information the user is interested in from the content state by the state determination and the history of the content viewing operation request, and determines the information. It generates as user information (output order reference) and stores it in the user information storage unit 12.

  Although not shown, the browser device 1 can be realized by a computer having a CPU such as a microprocessor, a recording unit such as a semiconductor memory or a magnetic disk, and a communication unit. The recording means stores a program that realizes the function of each component included in the browser device 1. The CPU reads the program to control the operation of the browser device 1, and the function of each component is recorded. Can be realized.

  Further, when voice output is performed by the voice output unit 3, the user can specify a part he / she wants to hear and output it again by operation. For example, when the user designates a repetitive operation while reading the content of the heading 1 by voice and the user request input unit 4 receives it as an output repeat request, the information is analyzed through the state determination unit 9 for content analysis. The content analysis unit 6 has a function of outputting the content being read again by instructing the content output unit 8 to output again the part of the content to be repeatedly output. Further, this operation may be performed by a voice operation such as, for example, “again” utterance. Further, as an operation for skipping unnecessary sentences, operations such as “pass” and “skip” are also conceivable. In this case, the user request input unit 4 inputs an output cancellation request, and the content analysis unit 6 instructs the content output unit 8 to output the next information after the current information, whereby the content output unit 8 outputs the information. Can be skipped.

  When the above operation is performed, the state determination unit 9 extracts how the content is output. For example, when screen display is performed, it is determined which part of the entire content is displayed, so if information on the display part is extracted or audio output is performed, It is determined which part of the content is being read out. Specifically, the information that the sentence “Keihan Riders wins Dantotsu” with the headline “Sports” is read out.

  The user information generation unit 10 estimates what information the user is interested in from the obtained information, and generates the information as user information. FIG. 3 shows an example of the processing procedure.

  First, the user information generation unit 10 extracts a sentence that is a target of output repeat (repetitive output) or output cancellation (output skip) (S101), and decomposes the extracted sentence into words (words) (S102). Then, each decomposed word is extracted (S103), and it is determined whether there is the same word in the user information stored in the user information storage unit 12 (S104). If the same word is in the user information (Yes in S104), points (evaluation value) are added to the word. on the other hand. If the same word is not in the user information (No in S104), the word is newly registered in the user information (S106). In the example of FIG. 3, it is assumed that the sentence “Keihan Riders wins the championship of Dantotsu” whose heading shown in FIG. 2 is the target of output repeat. In this case, information such as “sports” as the category and “Keihan” and “riders” as the attention words are extracted, and points are added to each. When an operation such as “pass” or “skip” is designated, it is determined that the user is not interested in the content information being output. If the extracted information, for example, the sentence that was output at the time of the skip operation is a sentence that the Democratic Party Governor ’s election is advantageous to Prime Minister Kobayashi, then “DLP”, “President”, “Election”, “Kobayashi”, “Prime Minister”, etc. Words are extracted, and those words are recorded in the user information as negative information.

  FIG. 4 shows an example of user information. In this example, the extracted words are ranked and registered by points (evaluation values) according to the extraction frequency. The rank is used as a priority (reference) in determining the sentence output order.

  When the user information is accumulated, the information is used in the subsequent content output. The content analysis unit 6 uses the user information as a reference for determining information that the user is interested in. The processing procedure is shown in FIG.

First, the content analysis unit 6 breaks down the content received by the content reception unit 5 into sentences (S201). Next, for each sentence, the sentence is decomposed into words (words) (S202), the decomposed words are extracted (S203), and it is determined whether there is the same word in the user information (S204). If there is the same word in the user information (Yes in S204), the points set for the word are added to the sentence, and the same process is repeated for the number of words. When the points are counted for each sentence as described above, the content analysis unit 6 rearranges the sentences in the descending order of the points according to the points of each sentence, and sets them in the output order (S206).
In the previous example, if there is a sentence including “Keihan” and “Riders”, the point of that sentence becomes higher, so the output priority of the sentence is raised. As a result, a sentence having a high point, that is, a sentence that is considered to be of high interest to the user can be displayed in the content output unit 8 or output in voice.

  As described above, according to the browser device according to the present embodiment, it is possible to perform operations by voice and the like, and it is possible to output contents in an appropriate order based on the accumulated user information. That is, by extracting and using user information, it is possible to preferentially output information desired by the user.

Embodiment 2. FIG.
FIG. 6 is an explanatory diagram illustrating an output example of the browser device 1 according to the second embodiment. Audio output is effective when the screen output is limited due to a small screen size or operational problems such as in a car navigation system. The content output unit 8 of the browser device 1 according to the present embodiment highlights the character portion being read aloud on the screen. In the upper example of the figure, the part being read is highlighted on the screen with thick characters. Moreover, you may make it change the color of a character. In the example below, only the part that is actually flowing is output like a telop.

  Generally speaking, there are many mistakes in reading aloud, however, by highlighting the part that is actually read out on the screen in this way, it is easier for the user to check the part of the reading error. , Can help understand the content. In other words, according to the browser device according to the present embodiment, it is possible to easily confirm a reading error at the time of voice output.

Embodiment 3 FIG.
Next, a browser device according to Embodiment 3 will be described.
As described in the first embodiment, in content analysis, if user information is used as information that the user is interested in, and there is a sentence including the corresponding word (such as “Keihan” or “Riders” in the previous example) The information for emphasizing the word itself is added, and the analysis information is output.

  FIG. 7 is an explanatory diagram illustrating an output example of the browser device according to the third embodiment. As shown in the figure, the content output unit highlights the words included in the user information based on the added information, or displays the area of the portion including the corresponding words on the screen. If it is an audio output, the corresponding sentence is read out.

Embodiment 4 FIG.
FIG. 8 shows an operation example of the browser device according to the fourth embodiment. Content analysis and normal output processing are the same as those up to the third embodiment.
In the present embodiment, the user outputs the next sentence including the word indicated in the user information, which is a word (word) included in the currently output sentence, for example, “next attention”. An operation request (non-output sentence output request) can be issued. When a request that can use user information is input in the form of “next attention”, the user request input unit 4 notifies the state determination unit 9 of the input. The state determination unit 9 searches for a word included in the currently output sentence and indicated in the user information, and outputs a non-output sentence including the word together with the searched word. To ask. The content analysis unit 6 determines whether or not there is a sentence including the word requested from the state determination unit 9 in the content received by the content reception unit 5, and if such a sentence exists, The sentence output order is determined so that the sentence is output with priority. As a result, the content output unit 8 can preferentially output the sentence.

  FIG. 8 shows the word “Keihan” included in the user information when a sentence “Keihan Riders wins the championship” is output and a user requests a non-output sentence output. In this example, an unoutputted sentence including “sentence of“ Keihan fan is mainly in Osaka ”” is output.

  As described above, according to the browser device according to the present embodiment, a browser operation based on attention information included in content can be performed based on user information. That is, according to the browser device according to the present embodiment, it is possible to appropriately output information desired by the user by preferentially outputting information that the user will pay attention to.

Embodiment 5 FIG.
FIG. 9 is a block diagram showing the basic configuration of the browser device according to the fifth embodiment. In the present embodiment, a user information editing unit 13 (output order reference editing unit) that edits user information through an operation from the user is provided.

  FIG. 10 shows a screen example when the user information editing unit 13 edits user information. In this example, user information as shown in FIG. 4 is displayed on the screen. When a word that is not intended by the user is registered, the user can delete the word by pressing the delete button from the list screen, or can register the word of interest with the “add” button. You can also modify the word content and frequency. The user information editing unit 13 edits the user information based on the addition / deletion / correction request from the user. The result of this editing is saved again as user information, and the updated information is used for the next content output.

  As described above, according to the browser device according to the present embodiment, it is possible to output information to be noticed more effectively by confirming and editing the information that the user will pay attention to by editing the information. .

Embodiment 6 FIG.
FIG. 11 is an explanatory diagram showing an example of user information stored in the browser device according to the sixth embodiment. When the user information is generated by the user information generation unit 10, the user information is generated together with the point information for each word and the location information indicating which Web page / site was used (URL column in the figure).
When the content analysis unit 6 performs analysis processing using user information, the content analysis unit 6 determines whether the location recorded in the user information matches the location to be displayed. By using, it is possible to output content with higher accuracy.

  Here, the features of the browser device described in the first to sixth embodiments will be described again below.

  The browser device according to the first embodiment can cut out a specific part in content in a Web browser that can mainly be operated and output by voice, and can repeatedly reproduce a part intended by a user. It is characterized by extracting and recording user information related to the user's interest from the operation and the content. Specifically, a content analysis unit that analyzes the inside of the content and generates analysis data for output, a content output unit that can reproduce each divided content, and can be specified for the divided content, And an operation unit capable of repetitive reproduction. Furthermore, a state determination unit that determines which content has been selected and a user information generation unit that generates word information included in the selected sentence as user information are provided.

  In addition to the first embodiment, the browser device according to the second embodiment is characterized in that specific information can be emphasized and output in content output.

  In addition to the second embodiment, the browser device according to the third embodiment is characterized by emphasizing using user information.

  In addition to the third embodiment, the browser device according to the fourth embodiment is characterized by adding a browse operation target using user information.

  The browser device according to Embodiment 5 includes a user information editing unit for editing user information.

  The browser device according to the sixth embodiment is characterized in that user information is managed in combination with location information.

1 is a block diagram showing a basic configuration of a first embodiment of the present invention. Explanatory drawing which shows the example of the content analysis of Embodiment 1 of this invention. Explanatory drawing which shows the process sequence of accumulation | storage of user information of Embodiment 1 of this invention. The block diagram which shows the example of the user information of Embodiment 1 of this invention. Explanatory drawing which shows the process sequence of the content analysis using the user information of Embodiment 1 of this invention. Explanatory drawing which shows the example of an information display at the time of reading-out of Embodiment 2 of this invention. Explanatory drawing which shows the example of information display of Embodiment 3 of this invention. Explanatory drawing which shows the operation example of Embodiment 4 of this invention. The block diagram which shows the basic composition of Embodiment 5 of this invention. Explanatory drawing which shows the example of operation of Embodiment 5 of this invention. Explanatory drawing which shows the example of the user information edit screen of Embodiment 6 of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Browser apparatus, 2 Display, 3 Audio | voice output part, 4 User request | requirement input part, 5 Content receiving part, 6 Content analysis part, 8 Content output part, 9 State determination part, 10 User information generation part, 11 Analysis information storage part , 12 User information storage unit, 13 User information editing unit.

Claims (9)

A content receiver that receives content including at least one sentence;
A content output unit that outputs sentences included in the content according to a predetermined output order;
A user request input unit for inputting a content viewing operation request from a user who uses the content;
An output order reference generation unit that generates an output order reference used to determine the output order of sentences based on a history of content viewing operation requests input by the user request input unit;
An output order determining unit that determines an output order of each sentence included in the content received by the content receiving unit using the output order reference generated by the output order reference generating unit. Processing equipment. The user request input unit includes:
As a content viewing operation request, at least one of an output repeat request for requesting to repeat output of a specific sentence included in the content received by the content receiving unit and an output cancel request for requesting to cancel output of a specific sentence You may enter
The output order reference generation unit
The information processing apparatus according to claim 1, wherein an output order reference is generated based on at least one of an output repeat request history and an output cancel request history. The content receiver is
Receive content that contains at least one word in the sentence,
The output order reference generation unit
When either an output repeat request or an output cancel request is input, the words included in the sentence that is the target of either the output repeat request or the output cancel request are extracted, and each word is selected according to the word extraction frequency. Set the evaluation value to, generate an output order criterion that shows the evaluation value for each word,
The output order determination unit includes:
For the sentence included in the content received by the content receiver, the word included in each sentence is compared with the evaluation value for each word indicated in the output order criterion to determine the output order of each sentence. The information processing apparatus according to claim 2. The content output unit
The information processing apparatus according to claim 1, wherein a sentence display and a voice output can be simultaneously performed as a sentence output, and a predetermined emphasis display is performed on a portion where the voice is output. The content output unit
4. The information processing apparatus according to claim 3, wherein, when outputting a sentence, predetermined highlighting is performed on a word whose evaluation value is indicated in an output order criterion. The user request input unit includes:
As a content viewing operation request, a non-output sentence that requests that the content output unit output a non-output sentence that includes a word that is included in the sentence that is being output and whose evaluation value is indicated in the output order criterion You may enter an output request,
The output order determination unit includes:
When a non-output sentence output request is input, the content output unit outputs a non-output sentence that includes a word that is included in the sentence that is being output and has an evaluation value set in the output order criterion. 4. The information processing apparatus according to claim 3, wherein the receiving unit extracts contents received from the content and prioritizes the output of the extracted unoutputted sentences to determine the output order of the sentences. The information processing apparatus further includes:
The information processing apparatus according to claim 1, further comprising: an output order reference editing unit that edits the output order reference generated by the output order reference generation unit based on an instruction from a user of content. The output order reference generation unit
4. The information processing apparatus according to claim 3, wherein an output order reference indicating an evaluation value for each word and identification information of a sentence from which a word is extracted is generated. A content receiving step of receiving content including at least one sentence;
A content output step of outputting sentences included in the content according to a predetermined output order;
A user request input step for inputting a content viewing operation request from a user who uses the content;
An output order reference generation step for generating an output order reference used to determine the output order of sentences based on the history of content viewing operation requests input in the user request input step;
An output order determining step for determining an output order of each sentence included in the content received by the content receiving step using the output order reference generated by the output order reference generating step. Processing method.

Images