JP2002077241A - Asynchronous voice data processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve voice quality in an asynchronous voice communication system. SOLUTION: The held data of a data-holding buffer 23, held after a voice signal received from an asynchronous communication network is decoded, are transferred to a data reproducing buffer 24. When the data read from the data reproducing buffer are outputted, the received data bytes to the data-holding buffer 23 are counted, and the output data bytes from the data-reproducing buffer 24 are counted. Then, a voice data reproducing speed is compared with a voice data-receiving speed, and when the voice data-receiving speed is faster than the voice data-reproducing speed, excessive received data are discarded. When the voice data-receiving speed is slower than the voice data-reproducing speed, by having the previously received voice data added, the generation of the slip of the voice data can be prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an asynchronous audio data processing system, and more particularly, to an asynchronous audio data processing system for improving sound quality deterioration caused by asynchronous operation.

[0002]

2. Description of the Related Art Conventionally, data synchronization between a data sending side device and a data receiving side device is established by extracting a synchronization clock from a communication line and establishing synchronization between opposing points using the extracted synchronization clock. The communication is realized by synchronizing the sampling clock and the line clock.

On the other hand, in recent years, there has been a strong demand for communication using a personal computer (PC), the Internet, or the like. However, when constructing a voice communication system using a TCP / IP network or a personal computer, which is currently spreading rapidly, clock reproduction, voice sampling, and synchronization of a line clock are generally performed. Can not do.

[0004] Regarding the line speed and asynchronous processing in the apparatus, for example, Japanese Patent Laid-Open No. 7-284077 discloses an ISD.
When transmitting and receiving PC-based device data in communication using N lines, control the loss and duplication of audio information that occurs because the amount of encoded audio data in the audio encoder is asynchronous with the transmission speed of the ISDN line. A synchronous slip processing technique is disclosed.

[0005]

SUMMARY OF THE INVENTION Japanese Patent Application Laid-Open No. 7-28407
In Patent Document 7, as described above, the control for the slip of the encoded data is performed, but the following slip caused by excess or deficiency of the PCM is not addressed. In other words, when the audio data receiving process and the playback process on the PC are asynchronous, slipping occurs due to the collapse or lack of the audio data, which causes problems such as deterioration of audio quality and normal reproduction failure. Not resolved.

Accordingly, an object of the present invention is to provide an asynchronous audio data processing system in which sound quality deterioration due to asynchronous is improved.

[0007]

In order to solve the above-mentioned problem, the asynchronous voice communication system according to the present invention employs the following characteristic configuration.

That is, according to the present invention, when audio data reception processing and reproduction processing by a PC are asynchronous, when the PCM audio data obtained by decoding is excessively or deficient and a slip occurs, If the decoding amount of PCM data is large, the data is discarded, and if the decoding amount of PCM data is smaller than the reproduction, the immediately preceding audio data is added to prevent audio data from slipping and ensure audio quality. I do. In this way, by monitoring the received audio data and the timing of the reproduction and controlling the reproduction of the audio,
The voice quality in the asynchronous voice communication system is improved.

More specifically, (1) In an asynchronous audio data processing system for reproducing audio data received via an asynchronous communication line, if the audio data reception speed is faster than the audio data reproduction speed, excessive reception data An asynchronous audio data processing method that discards audio data and prevents the occurrence of a slip in audio data by adding the audio data received immediately before when the audio data reception speed is lower than the audio data reproduction speed.

(2) The audio signal received from the asynchronous communication network is decoded and held in the data holding buffer, and the data held in the data holding buffer is transferred to the data playback buffer and read from the data playback buffer. In the asynchronous audio data processing method for reproducing and outputting the data, the received data bytes to the data holding buffer are counted, the output data bytes from the data playing buffer are counted, and the data holding buffer is counted based on both count values. An asynchronous audio data processing method for controlling data transfer from the data reproduction buffer to the data reproduction buffer.

(3) Based on a result of comparison between a difference between the received data byte count value and an output data byte value from the data reproduction buffer and a predetermined judgment reference value N, the data from the data holding buffer is read from the data holding buffer. The asynchronous audio data processing method according to the above (2), which controls data transfer to a reproduction buffer.

(4) The criterion value N is set based on a decrease in the quality of reproduced sound caused by reproduction of old data due to overwriting of data in an unreproduced area of the data reproduction buffer or lack of data. 3) Asynchronous audio data processing method.

(5) If the write position to the data reproduction buffer exceeds the reproduction position, the received data is discarded, and if the reproduction position exceeds the write position, the data is returned to the area for reproducing the immediately preceding audio data. The asynchronous audio data processing method of (2) to be written.

(6) The audio signal received from the asynchronous communication network is decoded and held in the data holding buffer, and the data held in the data holding buffer is transferred to the data playback buffer and read from the data playback buffer. An asynchronous audio data processing method for outputting data from a data holding buffer to the data reproduction buffer based on a write position and a reproduction position of the data reproduction buffer.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of an asynchronous audio data processing system according to the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 shows a personal computer (P) in an asynchronous voice data communication system according to an embodiment of the present invention.
It is a structural block of C).

In FIG. 1, the personal computer includes a data transmission processing unit 1 and a data reception processing unit 2. The data transmission processing unit 1 for PCM audio data receives, for example, an audio signal from a microphone from an input terminal A, converts the audio signal into digital data by an A / D conversion unit 11,
Is encoded by the socket transmitting unit 13
Is sent out from an output terminal C to an asynchronous communication network such as a TCP / IP network.

On the other hand, the data reception processing unit 2 converts the audio signal received from the asynchronous communication network via the reception terminal D by the socket reception unit 21 into a predetermined signal format.
After the data is decoded by the decoder 22, the data is held (saved) in the data holding buffer 23.

The data held in the data holding buffer 23 is transferred to a data reproducing buffer 24 under predetermined control, and the data read from the data reproducing buffer 24 is converted into an analog signal by a D / A converter 25. The signal is output to, for example, a speaker via the output terminal D.

In the present embodiment, the data input (data receiving speed) from the decoder 22 to the data holding buffer 23 and the D / A conversion unit 2
5 (data reproduction speed), and control of data transfer from the data holding buffer 23 to the data reproduction buffer 24 to control reproduction of PCM data. For this purpose, in the present embodiment, a reception counter unit 26 that counts the number of data bytes received by the data holding buffer unit 23, a reproduction counter unit 27 that counts the number of data bytes reproduced by the data reproduction buffer unit 24, A control unit that controls writing of the reception data held in the data holding buffer to the data reproduction buffer based on the count information from the counter unit.

As described above, the audio data is written into the data reproduction buffer 24, the digital data is converted into the analog data by the D / A converter 25, and the audio is output from the output terminal B. At the same time, the data reproduction counter 27 counts the number of bytes of the reproduced data. The control unit 28 controls the transfer and writing of the audio data saved in the data holding buffer 23 to the data reproduction buffer 24 based on the counter information of the data reception counter 26 and the counter information of the data reproduction counter 27.

Next, the flow of the received voice reproduction process and the control process of the received voice reproduction data after the decoding process of the received data in the decoder 22 shown in FIG. 1 will be described with reference to the flow chart of FIG. 2 and the configuration example of the buffer memory of FIG. I will explain while.

In FIG. 2, when data is received from the asynchronous communication network line, first, in step S1, audio data is decoded and stored in the data holding buffer 23, and the number of data reception bytes is counted by the data reception counter 26 (FIG. 2). Step S2). Next, reception is performed based on a comparison result of a difference DF between the data reception counter value by the data reception counter 26 and the data byte counter value by the data reproduction counter 27 that counts the number of bytes of data to be reproduced, and the determination reference value N. The writing and transfer of data to the reproduction buffer 24 are controlled (steps S3 to S6).

That is, the determination reference value N is set to a value necessary to avoid a decrease in the quality of reproduced sound caused by reproduction of old data due to overwriting of data in an unreproduced area of the data reproduction buffer 24 or reproduction of insufficient data. You.

For simplicity of understanding, for example, a simplified data reproduction buffer 24 as shown in FIG. 3 will be described. When the data reproduction buffer 24 is divided into three blocks and the criterion value N is set to 2, First, as shown in the normal operation, when the difference between the number of write data bytes and the number of reproduced data bytes is smaller than 2, and the number of write data bytes is larger than or equal to the number of reproduced data bytes, the sound quality is improved without destroying the data. Can be kept.
The case where the line speed and the reproduction speed are the same corresponds to this operation.

Next, if the difference between the number of write data bytes and the number of read data bytes is 2 or more and the number of write data bytes is larger than the number of read data bytes, as shown in the operation of overwriting the write position with the read position, Since the new audio data is overwritten on the M (2) area which is not used, the audio may be broken. Since the line speed of the received data is higher than the voice reproduction speed, the data amount is large and the reproduction cannot be performed in time. In such a case, in the present embodiment, a process of discarding the received data is performed.

Further, when the number of reproduction data bytes is larger than the number of write data bytes as shown in the operation where the reproduction position overtakes the write position, this means that the writing of new audio data is up to M (1) whereas the writing of new audio data is up to M (1). This is a case where the M (2) area to which no writing has been performed is to be reproduced. Since the data reproduced at this time is data that has been reproduced previously, the sound may be broken. Since the audio reproduction speed is faster than the line speed, the reproduction process is performed before the next data comes. In such a case, in the present embodiment, a process of minimizing a sense of incongruity of voice is performed by writing the immediately preceding voice data in an area to be reproduced.

The processing of steps S3 to S6 in FIG. 2 using the above-described determination and control will be specifically described. Step S
3, the determination result on the difference DF is 0 ≦ D
In the case of F ≦ N, the received data is written into the data reproduction buffer 24 in step S4 because the data reception and the reproduction process are balanced. Also, if the judgment result is DF> 0
In the case of, the data reception speed is higher than that of the reproduction process, so that the process shifts to step S5 to discard the received data. Further, when the result of the determination is DF <0, since the reproduction processing is faster than the data reception speed and the data amount is small, the flow shifts to step S6 to store the received data immediately before being held in the data reproduction buffer. Write. Step S4
After the processing of S6, the preparation for reproduction is completed.

When the sound reproduction process is performed, reproduction completion information is obtained in step S7, and in step S8,
The data reproduction counter is counted and the audio reproduction process is completed.

According to another embodiment of the present invention, while the basic configuration is as described above, in the determination method, the reproduction is performed by comparing the count value of the write data byte number and the count value of the reproduction data byte number. Instead of controlling the data, a method of controlling the data by constantly monitoring the write position and the reproduction position of the data reproduction buffer as shown in FIG. 3 can also be used.

The preferred embodiment of the asynchronous voice data processing system according to the present invention has been described in detail. However, it should be noted that such an embodiment is merely an example of the present invention and does not limit the present invention in any way. It will be readily apparent to those skilled in the art that various modifications can be made in accordance with the particular application without departing from the spirit of the invention.

[0032]

As described above, according to the present invention,
The audio data receiving process and the playback process on the PC are asynchronous,
If the audio data is slipped due to crushing or missing, if the data reception speed is faster than playback, if the received data amount is large, discard the data and if the data reception speed is slower than playback, Since the amount of received data is small, the immediately preceding audio data is added to prevent the occurrence of slips in the audio data, thereby ensuring the quality of the audio. Therefore, by monitoring the received audio data and the reproduction timing and performing audio reproduction control, an audio communication system in an asynchronous system for improving the audio quality is realized.

[Brief description of the drawings]

FIG. 1 shows a PC of an asynchronous audio data processing system according to the present invention.
FIG. 2 is a block diagram of a configuration on a (personal computer) side.

FIG. 2 is a flowchart showing a flow of a voice control method in the asynchronous voice data processing method according to the present invention.

FIG. 3 is a diagram for explaining an example of an audio data control determination method in the asynchronous audio data processing method according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Data transmission processing part 2 Data reception processing part 11 A / D conversion part 12 Encoder 13 Socket transmission part 21 Socket reception part 22 Decoder 23 Data holding buffer 24 Data reproduction buffer 25 D / A conversion part 26 Data reception counter 27 Data reproduction counter 28 Control unit

Claims (6)

[Claims] In an asynchronous audio data processing method for reproducing audio data received via an asynchronous communication line, if the audio data reception speed is higher than the audio data reproduction speed, excess received data is discarded and the audio data reproduction is performed. An asynchronous audio data processing method, wherein when the audio data reception speed is lower than the speed, the occurrence of audio data slip is prevented by adding the audio data received immediately before. 2. An audio signal received from an asynchronous communication network is decoded and held in a data holding buffer. The data held in the data holding buffer is transferred to a data reproduction buffer and read from the data reproduction buffer. In the asynchronous audio data processing system for reproducing and outputting data, while counting received data bytes to the data holding buffer, counting output data bytes from the data reproducing buffer, and from the data holding buffer based on both count values, An asynchronous audio data processing method, wherein data transfer to the data reproduction buffer is controlled. 3. A method for reproducing data from the data holding buffer based on a result of comparison between a difference between the received data byte count value and an output data byte value from the data reproduction buffer and a predetermined judgment reference value N. 3. The asynchronous audio data processing method according to claim 2, wherein data transfer to the buffer is controlled. 4. The determination reference value N is set on the basis of a decrease in quality of reproduced audio caused by reproduction of old data due to overwriting of data in an unreproduced area of the data reproduction buffer or lack of data. The asynchronous audio data processing method according to claim 3, wherein 5. When the write position to the data reproduction buffer exceeds the reproduction position, the received data is discarded, and when the reproduction position passes the write position, the immediately preceding audio data is written to the reproduction area. 3. The asynchronous audio data processing method according to claim 2, wherein: 6. An audio signal received from an asynchronous communication network is decoded and held in a data holding buffer, and the data held in the data holding buffer is transferred to a data reproduction buffer and read from the data reproduction buffer. An asynchronous audio data processing method for outputting data, wherein data transfer from the data holding buffer to the data reproduction buffer is controlled based on a write position and a reproduction position of the data reproduction buffer. .