KR100460411B1 - A Telephone Method with Soft Sound using Accent Control of Voice Signals - Google Patents

Abstract

The present invention relates to a method for improving the function of a telephone which is very widely used for communication in real life. The voice of the call from the other party is different. Because the other party is invisible, the person's feelings are offended by a rushing voice, a swearing voice, a mixed dialect, etc. The phone you need is a soft-sound (soft-sound) phone. When the receiver presses the soft-sound key (or certain key button) attached to the phone, the voice of the other party's voice does not go out of range. Invented to sound as.

Soft voice The phone analyzes the other party's voice information received through the phone, leaving the other party's semantic information intact, and adjusting the accent information indicating personality, so that the tone of the voice does not exceed a certain range so that it can listen smoothly and generously. It is an implementation.

In particular, related service workers who receive a large number of unspecified customers by telephone communication are stressed by the various types of voices of customers. Even in this case, the soft sound phone function calms the intonation of the customer's voice and can relieve some of the stress of professionals in the voice-related field.

Soft sound The phone can listen to the other party's words softly and calmly because it can select the soft sound call function according to the recipient's preference when the other person speaks quickly or speaks with a strong accent dialect. In addition, the phone with the soft sound function has a characteristic of inducing the voice to be quietly spoken by informing the invisible counterpart that the receiver is currently listening to through the side tone. Therefore, it is effective to calm and brighten the atmosphere of the telephone etiquette of the urgent and blunt society.

Description

A Telephone Method with Soft Sound using Accent Control of Voice Signals}

The present invention is to improve the method of making a telephone call in the Internet call, general telephone, mobile phone, etc. using the telephone network can be classified as voice conversion technology in the field of voice communication or audio signal processing.

Telephones connected to the existing telephone network currently used are transmitting the voice of the other party through the handset as shown in FIG. This method has the disadvantage that the invisible personality or the hardness contained in the other party's voice is transmitted to the receiver as it is, so that the receiver sometimes feels a lot of discomfort or stress. The present invention seeks to remedy this drawback.

The present invention newly proposes a method of calling a voice of the other party heard from the telephone receiver of a telephone with a soft sound (soft-sound) without strong intonation by applying digital speech processing technology. The human voice is generated by the vibration of the air from the lungs as it trembles in the vocal cords, and when this tremor is picked up through the saints, resonance occurs. The vocal vocal tremor and vocal habits of the voice represent the personality of the vocalists, and the resonance characteristics of the vocal tracts represent phonological information that conveys the meaning of the message. In this way, the resonance characteristic of the saints representing the meaning of the message is preserved as it is, and by adjusting the intonation information indicating the personality of the speaker as necessary, the invention invented by applying the principle that the voice is soft and sound to the telephone call.

1 is a principle of a soft sound telephone

2 is a voice analysis and synthesis scheme

3 is an example of a hardware configuration for intonation conversion processing

4 is an example of functional processing for a soft sound telephone;

[Basic configuration]

When we talk through a phone terminal such as a mobile phone or a landline phone, the sound wave signal is compressed by digital processing technology and transmitted to the other party's phone through a telephone repeater. The other side of the phone decompresses the digital signal, converts it into sound waves, and delivers it to our ears. As such, most phones have a high-performance computer chip that compresses or recovers voice signals, and provides various services such as text service.

The soft sound telephone adds the soft sound function 104 to the existing telephone function 102 as shown in FIG. When the phone rings and the handset is lifted, the other party's sound 107 is heard. If the other party's sound is urgent or indistinct, the soft sound-keyboard 101 is pressed. A soft sound-keyboard is a separate keyboard or is a way of pressing a random number-button. Pressing the soft sound-keyboard allows the other's voice to be controlled, for example, when the intonation is controlled (108), allowing the user to listen softly and accurately. On the other hand, when the soft sound-keyboard 101 is pressed once again, it returns to the state of the original intonation change of the other party's voice.

As shown in Fig. 2, the soft sound telephone adds an advanced processing function to make the voice sound smooth by adjusting the intonation information while maintaining the semantic information and tone characteristics by analyzing the voice signal processing in the computer chip built into the existing telephone. will be. That is, the voice is generated by the tremors of the vocal cords and the resonance in the throat. Using the principle of generating the voice, the voice characteristic is kept as it is (206). 207), a soft and gentle voice 208. The core technology of the soft sound phone automatically classifies the phonological information and the rhyme information indicating personality in the human voice (202) to preserve the personality and semantic information of the speaker (206) while simultaneously changing the intonation information (205). By increasing the soft characteristic 208 of the voice.

[Configuration of Hardware Device]

An example of a hardware device that receives an analog voice signal 300 from a microphone, a telephone line, or the like and processes the voice signal with a soft voice is illustrated in FIG. 3. The voice signal waveform 300 input in the analog form is amplified by the amplifier 301 and then passed through the low pass filter 302 to eliminate the aliasing effect. By passing through the analog-to-digital converter 304 that performs quantization and coding, the signal is converted into a digital signal in the form of a linear pulse code modulation (PCM) and input to a general purpose CPU or a digital signal processor (DSP).

When the signal is processed, the computer processor 304 may refer to a peripheral device 309 installed both inside and outside, and may also refer to the peripheral memory 305 to store input digital signals or processing results.

The digital signal that is processed into the intonation by the software or firmware by the CPU is converted into a sampled analog signal through the digital-to-analog converter 308. Passing this signal through the low pass filter 307 results in an analog signal from which quantization noise has been removed, and when amplified to an appropriate level (306), the signal becomes an analog signal 310 that can be heard through a telephone receiver, a speaker, or the like.

[Software Process]

A soft sound phone is a software that adds soft sound software or firmware to a CPU chip that functions as a traditional phone. When a telephone call is made, it is determined whether the soft sound key button (or any key button) is pressed, and if it is not pressed, general voice communication 420 is performed like the existing telephone. The soft sound key button is configured as a software toggle switch, which is turned on once by pressing it and turned off by pressing it again.

When the soft sound telephony function is started, the data sample 401 input from the analog-to-digital converter (ADC) is processed simultaneously in units of one frame. First, it is determined whether the data value in the current frame is the voiced sound interval, and if it is not the voiced sound interval (404), the occupancy ratio (Buffer Rate, BR) of the ring buffer is calculated. In order for the other party's voice to be heard softly, the other party's voice should be heard differently than the actual voice vocalization speed. The memory buffer required to wait for the processed data is called a ring buffer 409.

The occupancy rate of the ring buffer (BR) represents the time rate at which the data processed by the soft sound function is waited in the ring buffer. The occupancy ratio of the time when the current frame is a non-voiced sound and the waiting time in the ring buffer is determined (for example, 0.8 <BR). <1.2), the duration adjustment 416 may be performed to change the voice speed. This makes it possible to eliminate fluctuations in the uttering time caused when the soft sound function is performed. That is, only the voiced sound section 418 processes the data so that the sound is soft, but at this time, since the voice speed may be different from that of the original voice, the voice delay is adjusted in the non-voice sound section to eliminate the overall time delay. .

A method 403 for measuring whether the current frame is a voiced sound section or a non-voiced sound section has been proposed in the speech processing textbooks (Refs. 1, 3, 4). For example, the energy level can be easily measured. That is, if the average energy of the current frame exceeds a predetermined threshold and lasts 5 frames (100 ms), for example, this section becomes a voiced sound section.

If the data of the current frame is the voiced sound section 418, the soft sound function processing is performed on the data. The soft sound function detects the pitch intonation in the data of the current frame (406) and considers the pitch accent gradient (PAC) in units of frames, so that the PAC is out of the range of change (e.g. 1.5 times). Surface 419 changes the pitch intonation (407).

The change of the pitch intonation is processed in units of one block of voiced sound, which represents one block section of the voiced sound frame which is continuously detected. For a block of voiced sound, the pitch-inhibition change 407 is performed in units of pitch periods. For example, if the PAC exceeds a predetermined change range, the pitch change is performed to maintain the pitch period within a given maximum range. Many methods for changing the pitch period have been proposed so far (Refs. 1 and 2). For example, a pitch synchronous overlap and add (PSOLA) pitch change method for reconstructing a waveform by broadly segmenting a speech waveform in a pitch period unit in a time domain and then superimposing the changed waveform unit in a pitch period unit (Ref. 2).

At this time, it is necessary to accurately detect the pitch period, a number of methods have been proposed in recent years for the pitch period detection method of speech signals (Refs. 1 and 4). For example, the pitch detection is mainly used for the autocorrelation function, and there is a method for detecting the period of the repetitive waveform by calculating the correlation between adjacent voice waveforms (Ref. 1).

Since the processed data becomes unnatural and unnatural in the waveform, it is necessary to perform an energy amplitude change control 408 so that the change in amplitude naturally follows. For example, the change in energy amplitude is processed in units of pitch periods, and is performed by multiplying the waveform by the average energy amplitude of one pitch period.

The processed voice data is stored in a ring buffer (409), and output through a handset or a speaker in units of voice data through a digital-to-analog converter (DAC) according to the stored order (410). The function of the soft sound phone is handled in real time. That is, the processing of the soft sound telephony function 410 must be completed until the data of one frame is received from the analog-to-digital converter (ADC) 401 until the data of the next frame is received.

[references]

[1] Myung-Jin Bae, Sang-Hyo Lee, Digital Speech Analysis , Dong Young Publishers, 1998.

[2] Bae Myung-jin, Digital Speech Synthesis , Dong Young Publishing Co., 1998.

[3] Myung-Jin Bae, Digital Voice Coding , Dong Young Publishers, 1996.

[4] Rabiner and Schefer, Digital Signal Processing of Speech Signals , Prentice Hall, 1978.

Technically, the present invention is one of technologies that assist the five senses of human beings. As everyone ages, the sensational functions become aging and they hate the sound of large intonation changes. The soft sound phone is unique in that it is the practical use of welfare technology to complement these sensational functions. Therefore, its applicability is unique as a communication technology for the welfare state that can be provided to the elderly or the disabled.

And this soft sound phone is a key technology for building a society of emotion. Since society is highly advanced and humans are increasingly isolated, one-sided conversations are becoming very common, even in everyday phone calls. The atmosphere of this society can be changed through a soft sound phone. By changing the mood to speak softly and tenderly to the other person's urgent and changing one-sided voice, it can calm and calm the mood that can cause emotion.

In addition, the voice conversion technology applied to the soft sound phone is a practical invention technology that can be used as an auxiliary system of a professional shorthand for recording the contents of a conversation, and can be applied to a language learner to develop English listening ability. Therefore, this soft sound phone is a technology that can be utilized in various ways in real life, and its effect is very large.

Claims (1)

If you are talking to the other party using a telephone and the voice intonation of the other party is severe and urgent, you can use the soft sound function that converts the voice into a soft and emotional sound by pressing any key button. Implementing on a part of a chip or a CPU chip, extracts the feature of the voice signal and limits the pitch intonation information of the speaker within a specific range while maintaining the speaker's semantic and characteristic information. Detects the accent and adjusts the range of change to produce a soft voice, detects the change in the voiced accent so that it does not deviate from a certain range, or divides the duration of the utterance into voiced and non-voiced sounds, and then processes it differently. Implements voice conversion processing on the phone Telephone manner that allowed the soft sound Riga sounds so soft and affectionate.