JP2010093328A - Sound signal communication system, speech synthesis device, method and program for speech synthesis processing, and recording medium stored with the program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a sound signal communication system where conversation can be performed simultaneously among three or more sets of speech terminals. <P>SOLUTION: The sound signal communication system (100) includes a plurality of terminals (10, 20 and 30) having communication functions and an IP network (40) for inter-connecting the terminals, and allows the terminals to communicate sound signals each other in RTP communications. Each of the plurality of terminals has a speech synthesis means (120). The speech synthesis means is characterized in that the means includes an information extracting means (122), a transmission time determining means (123), a sound signal summing means (125), and a signal limiting means (126). The information extraction means extracts time information from a plurality of received sound signals. Based on the extracted time information, the transmission time determination means (123) determines the transmission time of the received signals. Based on the determined transmission time, the sound signal summing means (125) adds the plurality of sound signals. The signal limiting means (126) compares the value of the summed sound signals with a predetermined limitation value, and reduces the value of the summed sound signals to the limitation value when the value of the summed sound signals is larger. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a voice signal communication system capable of calling simultaneously with three or more call terminals, a voice synthesizer provided in the terminal, a voice synthesis processing method performed by the voice synthesizer, and a voice synthesis processing program for executing the method And a recording medium storing the program.

  Conventional telephone systems provide one-to-one calls using a telephone line switching network, but IP telephone systems that have become popular in recent years provide simultaneous calls by a large number of people using Internet lines. (For example, Patent Document 1).

  However, in a call service with a large number of people as described above, it is necessary for users (callers) of a plurality of call terminals to alternately speak. If the users of a plurality of call terminals speak simultaneously, a plurality of audio signals are overlapped in the call terminal on the receiving side, but if the sound pressure difference between these audio signals is large, it is called “buzz” Playback noise occurs. This is a phenomenon that occurs due to human auditory characteristics. Due to this noise, the user of the call terminal fails to hear the content of the call and further feels uncomfortable.

  In addition, since a plurality of audio signals must be processed at a time in order to talk simultaneously with a large number of people, the amount of data to be processed increases. However, since existing mobile phones do not have sufficient processing capability, there is a possibility that they cannot be reproduced as normal sound.

  As an alternative audio signal process, there is a method of processing and outputting a plurality of audio signals in series in the order received. However, this processing method delays the playback timing of the audio signal, making smooth conversation difficult.

On the other hand, Patent Document 2 discloses an audio mixing technique that compares absolute amplitude values of a plurality of audio signal packets received at the same time in units of samples and selectively processes only a sample having the largest amplitude.
JP 2008-172420 A JP2005-151044

  The technology disclosed in Patent Document 2 is premised on the theory that “a small sound is drowned out by a loud sound”, and at the same time, selectively using only the loudest voice among the received voices, Other voices are discarded. When a conversation is actually carried out between three or more parties using this technology, only the voice of a specific caller is always heard, or the caller's voice is suddenly switched, so that a normal conversation is performed. Is difficult. Furthermore, when switching to any one of a plurality of sounds, if there is a large difference in sound pressure between these sound signals, there remains a problem that reproduction noise called “buzz” occurs.

  In addition, the inventors of the present invention have made extensive studies to solve the above problems, and as a result, a plurality of sound signals can be reproduced with a sufficient sound quality if a plurality of sound signals are simply digitally added to generate a composite signal. Confirmed the facts that can be made.

  However, the audio signal arrives at the receiving terminal with a delay from the time of transmission, depending on the processing structure and propagation path of the receiving terminal. Furthermore, audio signals sent from terminals of different transmission sources have different delay times. Therefore, if the simple addition is performed based on the time when the audio signal is received, there is a possibility that normal signal reproduction cannot be performed due to the difference in delay time between the audio signals.

  The present invention was devised in view of the above inconveniences, and an object of the present invention is to provide a voice signal communication system capable of simultaneous conversation between three or more call terminals, and a voice included in the terminals of the system. A synthesis device, a speech synthesis processing method performed by the speech synthesis device, a speech synthesis processing program for executing the method, and a recording medium storing the program.

  Another object of the present invention is to provide a speech synthesis system capable of synthesizing a plurality of received speech signals with high quality without imposing an excessive burden on the speech processing unit of the call terminal, and the system of the system. To provide a speech synthesizer provided in a terminal, a speech synthesis processing method performed by the speech synthesizer, a speech synthesis processing program for executing the method, and a recording medium storing the program.

  Still another object of the present invention is an audio signal communication system for synthesizing a plurality of received audio signals, and an audio signal communication system capable of correcting a difference in delay time generated between the audio signals, and the system A speech synthesizer included in the terminal, a speech synthesis processing method performed by the speech synthesizer, a speech synthesis processing program for executing the method, and a recording medium storing the program.

The invention of claim 1 devised to solve the above-described problem comprises a plurality of terminals having a communication function and an IP network interconnecting the plurality of terminals, and an audio signal between the plurality of terminals. An audio signal communication system capable of RTP communication,
Each of the receiving units of the plurality of terminals has speech synthesis means,
The speech synthesis means
(A) information extracting means for extracting time information from time stamps of RTP headers of a plurality of received audio signals;
(B) based on the extracted time information, transmission time determining means for obtaining a transmission time of the received signal;
(C) audio signal adding means for adding the plurality of audio signals based on the determined transmission time;
(D) comparing the value of the added audio signal with a predetermined limit value, and if the value of the audio signal is larger, signal limiting means for reducing the value of the audio signal to the limit value;
It is provided with.

Invention of Claim 2 created in order to solve the said subject is a speech synthesizer which can synthesize | combine the audio | voice signal received from the several transmission source via the RTP communication system containing an IP network in real time,
(E) information extracting means for extracting predetermined time information from time stamps of RTP headers of the plurality of received audio signals;
(F) a transmission time determining means for obtaining a transmission time of the received signal based on the extracted time information;
(G) audio signal adding means for adding the plurality of audio signals with the obtained time as a reference;
(H) audio output means for outputting the added audio signal;
It is characterized by providing.

  Invention of Claim 3 created in order to solve the said subject WHEREIN: In the speech synthesizer of Claim 2, the value of the audio | voice signal added by the said audio | voice signal addition means is compared with a predetermined | prescribed limit value, The said audio | voice is compared. In the case where the value of the signal is larger, the apparatus further comprises signal limiting means for reducing the value of the audio signal to the limit value.

Invention of Claim 4 created in order to solve the said subject is the audio | voice synthesis | combination processing method which synthesize | combines the audio | voice signal received from the some transmission source via the RTP communication system containing an IP network in real time,
(I) an information extracting step of extracting time information from time stamps of RTP headers of the plurality of received audio signals;
(J) a transmission time determining step for obtaining a transmission time of the received signal based on the extracted time information;
(K) an audio signal addition step of adding the plurality of audio signals based on the obtained time;
(L) an audio signal output step for outputting the added audio signal;
It is characterized by including.

  Invention of Claim 5 created in order to solve the said subject is the audio | voice synthesis processing method of Claim 4, The said audio | voice signal added between the said audio | voice signal addition step and the said audio | voice signal output step. And a predetermined limit value. If the value of the audio signal is larger, a signal limiting step of reducing the value of the audio signal to the limit value is further included.

  Invention of Claim 6 created in order to solve the said subject is an electronic circuit which performs the speech synthesis processing method of Claim 4 or 5.

  Invention of Claim 7 created in order to solve the said subject is a program which makes a processing apparatus perform the speech synthesis processing method of Claim 4 or 5.

  Invention of Claim 8 created in order to solve the said subject is a computer-readable medium which stored the program of Claim 7.

  The voice signal communication system according to claim 1, the voice synthesizer according to claim 2, and the voice synthesis processing method according to claim 4 synthesize a plurality of voice signals by simply digitally adding them. Therefore, the calculation is simple and no excessive load is generated in the processing unit of the terminal. Therefore, it can be suitably used even with a device having a relatively low computing capability such as a mobile phone.

  Further, the addition of the audio signal is performed in a state where the audio signal is arranged on the time axis with reference to the transmission time after the transmission time is grasped by referring to the header portion of the audio signal. As a result, the difference between the delay times of the audio signals is corrected, and the audio signals can be synthesized with high accuracy.

The voice signal communication system according to claim 1, the voice synthesizer according to claim 3, and the voice synthesis processing method according to claim 5, wherein the value of the synthesized voice signal is compared with a predetermined limit value, and the value of the voice signal is greater. If it is larger, the value of the audio signal is reduced to the limit value.
When a large number of audio data are simply added, the value of the added data becomes too large and exceeds the allowable input value of the terminal, and as a result, the reproduced sound quality may be impaired. Therefore, such a problem is prevented by providing a limiter in the composite data.

  In the sixth to eighth aspects, the speech synthesis processing of the present invention is provided in the form of an electronic circuit, a program, and a computer readable medium, respectively.

  According to the present invention, it is possible to provide an audio signal communication system for multi-party calls that corrects a difference in delay time between received audio signals, avoids complicated calculation processing of audio signals, and can reproduce high-quality audio. It has become possible. Furthermore, it is also possible to provide a speech synthesizer for a call terminal used in the system, a speech synthesis processing method performed by the speech synthesizer, a speech synthesis processing program for executing the method, and a recording medium storing the program. It has become possible.

A speech synthesis system according to an embodiment of the present invention will be described below with reference to the accompanying drawings.
FIG. 1 is a schematic diagram of an audio signal communication system 100 according to an embodiment of the present invention. The audio signal communication system 100 includes three terminals 10, 20, 30 and an IP (Internet Protocol) network 40. Terminals 10, 20, and 30 are configured by existing telephone devices such as telephones, portable terminals having a telephone function, or computers. The telephone voice is captured as an analog signal by an input unit (such as a microphone), and A / D converted to digital data. From the output unit (speaker, headphone terminal, etc.) as an analog signal after D / A conversion of the digital signal input from the other terminal via the IP network 40. A receiving unit for outputting. Moreover, although three terminals are shown in FIG. 1, the number of terminals may be any number of two or more. The IP network 40 is for interconnecting the terminals 10, 20, and 30 so as to allow communication. The network may be configured by wireless, wired, or a combination thereof.

The audio signal communication system 100 described above is a multi-party call system in which calls can be made simultaneously by the terminals 10, 20, and 30. For example, voices input to the input unit of the terminal 10 are output almost simultaneously from the output units of both the terminals 20 and 30. Conversely, the voices that are simultaneously input to both the input units of the terminals 20 and 30 are output almost simultaneously from the output unit of the terminal 10. The system is a serverless system that does not control the transmission right of each terminal.
In this embodiment, the audio signal communication system 100 conforms to SIP (session initiation protocol) when a session is established or terminated, and conforms to RTP (real-time transport protocol) when an audio signal is communicated.

  FIG. 2 shows the structure of the RTP header portion of the RTP packet constituting the audio signal handled by the audio signal communication system 100 according to the present embodiment. As will be described in detail later, the voice data synthesis unit 120 (see FIG. 3) of the terminals 10, 20, and 30 of the voice signal communication system 100 is configured with the order number, time stamp, and synchronization transmission source (SSRC) identifier of the header part. Audio data processing is performed using three types of information. Therefore, before describing the detailed functions of the terminals 10, 20, and 30, these header information will be briefly described.

[Sequence number]
One audio data is transmitted in a plurality of packets, and this sequence number is information indicating the number of packets in the audio data. The initial value is random, and the sequence number is incremented by 1 every time one packet is sent.
[Time stamp]
Information indicating the sampling time of the first byte of the packet. In this embodiment, it is used to specify the transmission time of received packet data.
[Synchronous transmission source (SSRC) identifier]
Source identifier. This is used to specify the source terminal of the received packet data.

  Next, the audio data processing unit of the terminals 10, 20, and 30 will be described with reference to FIG. These three terminals all have the same voice data processing function. The functions of this processing unit are roughly divided into a packet data transmission / reception unit 400, a reception unit 140, a transmission unit 200, and a communication session control unit 300.

  The packet data transmission / reception unit 400 is connected to the IP network 40 and performs establishment / termination of a session between the terminals 10, 20, and 30, and transmission / reception of voice data.

  The receiving unit 140 includes an audio data receiving unit 110, an audio data synthesizing unit 120, and an audio data output unit 130. The audio data receiving unit 110 is a functional element for receiving PCM format audio data sent from the packet data transmitting / receiving unit 400 and demodulating it into digital data for processing in a predetermined format. The voice data synthesizing unit 120 is a functional element for synthesizing transmission signals from a plurality of terminals received from the voice data receiving unit 110 with reference to their transmission times, and is mainly composed of an arithmetic unit such as a processor. . The audio data output unit 130 is a functional element for receiving the signal synthesized by the audio data synthesis unit 120 and outputting the signal to the outside, and includes a D / A converter, an analog circuit, a speaker, a headphone terminal, and the like.

  The transmission unit 200 includes an audio data input unit 210 and an audio data transmission unit 220. The voice data input unit 210 has a function of generating a digital data signal based on voice uttered by a terminal user, and includes a microphone, an analog circuit, an A / D converter, and the like. The audio data transmission unit 220 is a functional element that converts data received from the audio data input unit 210 into a compression code (PCM signal) using a PCM codec, adds an RTP payload, and sends the data to the packet data transmission / reception unit 400. It consists of a computing unit such as a processor.

  The communication session control unit 300 is a functional element that controls the session establishment / termination and data exchange timing of the packet data transmission / reception unit 400, and is mainly composed of an arithmetic unit such as a processor.

  Among the above functional elements, those other than the audio data synthesis unit 120 of the receiving unit 140 are functional elements known in the art, and thus detailed description of these functions is omitted. Therefore, only the function of the voice data synthesis unit 120 will be described in detail below.

  A detailed function of the voice data synthesis unit 120 will be described with reference to FIG. The voice data synthesis unit 120 includes a data reception unit 121, a header information extraction unit 122, a time difference determination unit 123, a clock unit 124, a data addition and restriction unit 125, a data output unit 126, and an output control signal generation unit. 127, a header information management unit 128, and an audio data buffer unit 129.

[Data receiving unit 121]
When the data receiving unit 121 receives audio data in units of packets from the audio data receiving unit 110 (see FIG. 3), the data receiving unit 121 transfers the packets to the data addition and restriction unit 125. At the same time, the clock unit 124 is accessed to obtain the current time, and the time is recognized as the reception time R1 of the packet. Further, when the received packet is the first packet after the session is established from the transmission source terminal, the information of the RTP bed portion of the packet is passed to the header information extraction unit 122.

  In addition to the above processing functions, the data receiving unit 121 also controls the output timing of the processed packet. Specifically, when the first packet after session establishment is received, the output control signal generation unit 127 is instructed to generate an output control signal. Further, when the clock unit 124 is periodically accessed to obtain the current time, the elapsed time since the reception of the latest packet is obtained, and the next packet is not input even after the preset period Tp has passed. Assumes that the communication has ended, and instructs the output control signal generation unit 127 to stop the output control signal. Note that the period Tp is not limited to a specific value, and is preferably determined as appropriate by the terminal setter based on the required specifications. Moreover, it is good also as a variable value so that a user can determine an optimal value.

[Header information extraction unit 122]
The header information extraction unit 122 extracts predetermined information (sequence number, time stamp, synchronous transmission source (SSRC) identifier) from the information of the RTP header part received from the data reception unit 121, and sends the information to the time difference determination unit 123. send.

[Time difference determination unit 123]
The time difference determination unit 123 refers to the information (sequence number, time stamp, synchronous transmission source (SSRC) identifier) of the RTP header part sent from the header information extraction unit 122, and first, based on the synchronous transmission source (SSRC) identifier. The transmission source of the packet received by the data receiving unit 121 is identified and used as transmission source information. Next, the transmission time T1 of the packet is obtained based on the time stamp, and further, a time difference Td between the transmission time T1 and the previously recognized reception time R1 is obtained. Here, time difference Td = reception time R1−transmission time T1. This time difference Td is determined by the time involved in the signal processing of the terminal on the transmission side, the propagation delay of the transmission line, the packet queuing delay in the router, and the like. Since this value is almost constant when the transmission source terminal and the transmission path are the same, in this embodiment, only the first packet of the transmission source terminal after session establishment is obtained.

  Next, the time difference determination unit 123 stores, in the header information management unit 128, the packet sequence number of the RTP header part sent from the header information extraction unit 122, the transmission source obtained by the above method, and information about the time difference Td. To do.

[Clock part 124]
The clock unit 124 is a functional element that provides the current time to the data receiving unit 121, and this current time is used to specify the reception time R1 and the elapsed period Tp.

[Data addition and restriction unit 125]
When the data addition / limitation unit 125 receives a packet from the data reception unit 121, it first accesses the header information management unit 128 to acquire the time difference Td of the packet. Then, the transmission time T1 is obtained from the previously obtained reception time R1 and the time difference Td. Then, the payload (voice data) of the received packet is stored in the area corresponding to the transmission time T1 in the voice data buffer unit 129, but before that, it is confirmed whether the data is already stored in the corresponding storage area. To do. If data has already been stored, the data and the data to be stored are added and stored in the area. The addition method used here may be simple addition.

With reference to FIG. 5, a supplementary explanation will be given regarding the data addition method performed by the data addition and restriction unit 125.
Data A and data B on the left side of the figure are waveforms of audio data input to the data addition and restriction unit 125 from different transmission source terminals. The transmission time T1 of data A is tA, and the transmission time T1 ′ of data B is tB. Here, the difference between tA and tB is twice the sampling period of these waveforms. In this case, both data are aligned on the time axis with reference to the respective transmission times T1 and T1 ′, that is, the waveform of data B is shifted to the right by two samples with respect to the waveform of data A. The samples at the same time are added together. The waveform of the synthesized data thus obtained is the right waveform.
As described above, the data addition and restriction unit 125 has a concept of a time axis, and performs addition in a state where input data is arranged on the time axis with reference to the transmission time T1.

It should also be noted that if packets are simultaneously received from a large number of terminals, all the payloads of these packets are added, resulting in a considerably large data value finally obtained. If this data is output to the processing unit in the subsequent stage in that state, the allowable input may be distorted, and the quality of the reproduced sound may be significantly impaired, or the subsequent processing unit may be damaged. Therefore, in order to prevent the above problem, the data addition and restriction unit 125 changes this value to the upper limit value when the data handled becomes a value larger than the reference value (hereinafter, this process is referred to as a data restriction process). Called).
Here, the upper limit value Th serves as a limiter, and the upper limit value Th may be set to the upper limit of the effective numerical value range of the PCM data, for example. In this case, the upper limit value Th is set to 255 when the payload is 8 bits, and is set to 32767 when the payload is 16 bits.

[Data output unit 126]
When the data output unit 126 receives the output control signal from the output control signal generation unit 127, the data output unit 126 reads the data stored in the audio data buffer unit 129 in accordance with the output control signal, and outputs the data to the subsequent audio data output unit 130 (see FIG. 3). To do.

[Output control signal generator 127]
This is a functional element that generates an output control signal in response to a command from the data receiving unit 121 and outputs the output control signal to the data output unit 126. When the output of audio data is started, an output control signal is generated, and when the output of audio data is stopped, the output control signal is stopped.

[Header information management unit 128]
This is a functional element for storing and managing information (transmission source, sequence number, time difference Td) related to received packets, and includes a register, a memory, a hard disk, and the like.

[Audio data buffer unit 129]
This is a functional element for temporarily storing data to be processed by the data addition and restriction unit 125, and includes a register, a memory, a hard disk, and the like. This buffer unit may be prepared exclusively, or a jitter buffer for correcting data fluctuation may be used. The storage area inside the buffer unit has a concept of a time axis, and the stored data is managed in association with the transmission time T1.

  FIG. 6 shows an example of a waveform of audio data output when two audio data are input in the audio data synthesis unit 120 having the above functions. In this figure, waveform 1 and waveform 2 are waveforms that are input to microphones or the like of terminals of two different transmission sources, and waveform 1 is a waveform that is generated when a person repeatedly utters “A” at a constant cycle. Yes, waveform 2 is a sine wave waveform of a 495 Hz frequency sound generated by the transmitter. Waveform 3 is a waveform that is synthesized from waveform 1 and waveform 2 and is output from the speaker of the receiving terminal. As shown in this waveform, the waveform 1 and the waveform 2 are correctly synthesized over the entire period, and there is no place where the level of the sound that causes “buzzy” noise suddenly changes.

The function of the audio signal communication system 100 according to the embodiment of the present invention has been described above. Next, a procedure of data synthesis processing performed by the voice data synthesis unit 120 in each terminal when the terminals 10, 20, and 30 communicate with each other in this system will be described.
The data synthesis process is roughly divided into two processes, a data addition process and a data output process. Therefore, these processes will be described in order. FIG. 7 shows the flow of data addition processing, and FIG. 8 shows the flow of data output processing.

[Data addition processing]
This will be described with reference to the flowchart of FIG.
First, the terminals 10, 20, and 30 shown in FIG. Since the audio signal communication system 100 according to the present embodiment does not have a server that performs communication control as illustrated, each terminal performs a session with its own terminal in addition to other terminals. By establishing the session, the three terminals can exchange voice data with each other in real time.

  After the session is established, any one of the terminals 10, 20, and 30 (receiving terminal) receives voice data packets from the other two terminals (transmitting terminals) via the IP network 40. This packet passes through the packet data transmission / reception unit 400 and the audio data reception unit 110 of the transmission terminal, and is input to the audio data synthesis unit 120. The packet received by the data receiving unit 121 in the voice data synthesizing unit 120 is received (step S10).

  The data receiving unit 121 refers to the synchronous transmission source (SSRC) identifier and the sequence number in the header of the received packet, and confirms whether this packet is the first packet after the start of the session from the transmission source terminal ( Step S11). In the case of the first packet (“YES” in step S11), the procedure proceeds to step S12, and in other cases (“NO” in step S11), the procedure proceeds to step S16.

If the received packet is the first packet (“YES” in step S11), the data receiving unit 121 accesses the clock unit 124, acquires the current time, and sets this as the reception time R1 (step S12). Further, the header information of the packet is passed to the header information extraction unit 122. The header information extraction unit 122 extracts information of a sequence number, a time stamp, and a synchronous transmission source (SSRC) identifier from the received header part information (step S13). Then, a time difference Td (R1−T1) is obtained based on the transmission time T1 of the data packet obtained from the time stamp and the reception time R1 obtained previously (step S14). Next, information on the transmission source, the packet order, and the time difference Td is stored in the header information management unit 128 (step S15).
It should be noted that the processing of steps S12 to S15 is performed only for “YES” in step S11, that is, only for the first packet from the transmission source terminal after the session starts.

  Next, in the process of step S16, the data addition and restriction unit 125 receives the packet from the data reception unit 121, and then accesses the header information management unit 128 to acquire the time difference information Td of the received packet. Next, the transmission time T1 of the received packet is obtained from the reception time R1 obtained in advance and the obtained time difference information Td (step S16). Then, the audio data buffer unit 129 is accessed to check whether data is already stored in the area corresponding to the transmission time T1 (step S17).

If data has already been stored (“YES” in step S17), the stored data is read out, and this data and the data being processed are arranged on the time axis with reference to their transmission time T1. Add (step S18) and proceed to step S19.
On the other hand, if there is no data (“NO” in step S17), the process proceeds to step S19.

Next, the value of the data obtained by the above procedure is compared with a predetermined limit value Th (step S19). If there is a value exceeding the limit value Th (“YES” in step S19), the value is changed to the limit value Th (step S20). If it does not exist (“NO” in step S19), no data change is performed.
Then, the corresponding data is stored in an area corresponding to the transmission time T1 of the data in the audio data buffer unit 129 (step S21).

[Data output processing]
This will be described with reference to the flowchart of FIG. Note that this data output processing is performed in parallel with the data addition processing described above.
First, when a session starts, the data receiving unit 121 of the receiving terminal causes the output control signal generating unit 127 to generate an output control signal, and causes the data output unit 126 to start an output operation (step S50). The data output unit 126 accesses the audio data buffer unit 129, extracts the data stored by the data addition and restriction unit 125 in the above-described data addition process (step S51), and outputs the data to the audio data output unit 130 (see FIG. 3). (Step S52). At the same time, the data receiving unit 121 periodically obtains the current time from the clock unit 124 and monitors the elapsed time after receiving the latest packet (step S53). When the elapsed period is within the range of the predetermined period Tp (“NO” in step S54), the procedure of steps S51 to S54 is repeated thereafter.

On the other hand, when the elapsed period exceeds the predetermined period Tp (“YES” in step S54), the receiving terminal considers that the reception of data has ended. Specifically, the data receiving unit 121 causes the output control signal generation unit 127 to finish generating the output control signal (step S55). When the data output unit 126 stops receiving the output control signal from the output control signal generation unit 127, the data output unit 126 stops the data output operation (step S56). Thereafter, the receiving terminal returns to the initial state of the session start and waits for the reception of the voice data packet.
The above is a series of procedures for synthesizing audio data in the audio signal communication system 100 according to the embodiment of the present invention.

  Note that the functions of the audio signal communication system 100 according to the embodiment of the present invention are not limited to specific hardware resources or software processing. That is, as long as the voice data synthesizing unit 120 of the terminals 10, 20, and 30 according to the embodiment of the present invention can realize the function, any hardware (electronic circuit, etc.), software (program), or a combination thereof can be used. May be used.

  When the speech signal synthesis method according to an embodiment of the present invention described above is implemented as a program, the program is downloaded from an external server or the like to an information processing apparatus that executes the method, or a computer-readable medium It is preferably distributed in the form. Examples of the computer readable medium include a CD-ROM, DVD, magnetic tape, flexible disk, magneto-optical disk, and hard disk.

  As mentioned above, although this invention was demonstrated using embodiment shown in drawing, these are only an illustration and those skilled in this technical field can variously be within the range which does not deviate from the range and the meaning of this invention. It will be understood that modifications and variations are possible. Accordingly, the scope of the invention should not be determined by the described embodiments, but by the technical spirit described in the claims.

1 is a diagram showing a configuration of an audio signal communication system 100 according to an embodiment of the present invention. It is a figure which shows the structure of the header part of RTP data with which communication of the audio | voice signal communication system 100 is based. 3 is a block diagram showing functions of audio data processing units of terminals 10, 20, and 30 of the audio signal communication system 100. FIG. It is a block diagram which shows the function of the audio | voice data synthesis | combination part 120 of FIG. FIG. 6 is a diagram for explaining a data addition method of a voice data synthesis unit 120. It is an example of the input / output waveform of the terminals 10, 20, and 30. It is a flowchart which shows the data addition process of the speech synthesis process which the speech data synthesis part 120 performs. It is a flowchart which shows the data output process of the speech synthesis process which the speech data synthesis part 120 performs.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 terminal 20 terminal 30 terminal 40 IP network 100 audio | voice signal communication system 120 audio | voice data synthetic | combination part 121 data reception part 122 header information extraction part 123 time difference determination part 124 clock part 125 data addition and restriction part 126 data output part 127 output control signal generation 128 Header information management unit 129 Audio data buffer unit

Claims (8)

A voice signal communication system comprising a plurality of terminals having a communication function and an IP network interconnecting the plurality of terminals, and capable of RTP communication of voice signals between the plurality of terminals,
Each of the receiving units of the plurality of terminals has speech synthesis means,
The speech synthesis means
Information extracting means for extracting time information from time stamps of RTP headers of a plurality of received audio signals;
Based on the extracted time information, transmission time determination means for obtaining a transmission time of the received signal;
Audio signal adding means for adding the plurality of audio signals based on the determined transmission time;
A signal limiting means for comparing the value of the added audio signal with a predetermined limit value, and if the value of the audio signal is greater, reducing the value of the audio signal to the limit value;
An audio signal communication system comprising: A speech synthesizer capable of synthesizing speech signals received from a plurality of transmission sources via an RTP communication system including an IP network in real time,
Information extracting means for extracting predetermined time information from time stamps of RTP headers of the plurality of received audio signals;
Based on the extracted time information, transmission time determination means for obtaining a transmission time of the received signal;
Audio signal adding means for adding the plurality of audio signals based on the determined time;
Audio output means for outputting the added audio signal;
A speech synthesizer comprising:   A signal limiting unit that compares the value of the audio signal added by the audio signal adding unit with a predetermined limit value, and reduces the value of the audio signal to the limit value when the value of the audio signal is larger. The speech synthesizer according to claim 2, further comprising: A voice synthesis processing method for synthesizing voice signals received from a plurality of transmission sources via an RTP communication system including an IP network in real time,
An information extracting step of extracting time information from time stamps of RTP headers of the plurality of received audio signals;
A transmission time determining step for obtaining a transmission time of the received signal based on the extracted time information;
An audio signal addition step of adding the plurality of audio signals based on the determined time;
An audio signal output step for outputting the added audio signal;
A speech synthesis processing method comprising: Between the audio signal addition step and the audio signal output step,
Comparing the value of the added audio signal with a predetermined limit value, and when the value of the audio signal is larger, further includes a signal limiting step of reducing the value of the audio signal to the limit value. The speech synthesis processing method according to claim 4, wherein:   An electronic circuit that performs the speech synthesis processing method according to claim 4.   A program for causing a processing device to execute the speech synthesis processing method according to claim 4 or 5.   A computer-readable medium storing the program according to claim 7.

Images