JPH03133233A - Voice packeting unit - Google Patents

Abstract

PURPOSE:To attain the processing of a first detection of a voiced part and the detection of packet abort with the same control at a reception side by adding a sequential number including a packet number of a silent period to a transmission packet. CONSTITUTION:A sequential number for each packet is noticed from a header analyzing section 7b to a sequential number monitoring section 7f, the sequential number monitoring section 7f monitors it to detect the voiced period and the silent period depending on the discontinuous of sequence number and when packet abort takes place, the period is processed as the silent period. A transmission side generates a transmission packet based on voice/silent information for each packet unit from a voice/silent detection section and sends the result. In this case, the sequential number from 0 to 15 including the silent period is assigned sequentially to the sequential number of the sent packet.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application] The present invention relates to an audio packetization device that encodes an audio signal and packetizes and transmits the data of the compressed voice portion.

[Conventional technology]

Figure 4 is a block diagram of a conventional audio packetization device;
The figure is a block diagram of the packet disassembly processing section (7). Note that the conventional example is based on Japanese Patent Application Laid-Open No. 63-257367.

In the figure, (1) is an A/D conversion unit that converts an analog audio signal into a digital signal, (2) is an encoding compression unit that encodes the audio signal and compresses information, (3) is a speech/silence detection unit, (4) is a packet assembly unit, (5) is a line control unit that sends packetized data to the line or receives packetized data from the line, (6) is a relay line, (7) is a packet separation unit, (7a) (7b) is a header analysis unit that analyzes the separated header information; (7c) stores audio data and outputs it to the decoding/expansion unit (8). (7d) is a playback timing control unit that controls the output timing from the fluctuation absorption buffer (7C); (7e) monitors the packet sequence number to detect whether or not a packet is discarded. A packet discard monitoring section, (8) a decoding/expanding section that decodes and expands received audio data, and (9) an A/D converting section that converts a digital signal into an analog signal.

Next, the operation will be explained. An analog voice signal from a telephone is digitized by an A/D converter (1) and then data compressed by an encoder and compressor (2). At the same time, the sound/silence detection unit (3) determines the sound/silence of the audio for each transferred packet, and only the sound part is packetized by adding header information in the packet assembling unit (4). be done. Furthermore, the packetized data is transferred to the 1-line controller (
5) is sent to the network via relay port n(6) according to the specified format (X, 25etcl).Packets received from the network side via the line control unit (51) are sent to the packet disassembly processing unit (7). After decomposition, the decoding and decompression section (
8), decoding processing for data expansion is performed, and further D/
The A converter (9) converts it into an audio analog signal and outputs it to the telephone.

In the packet disassembly processing of the audio packet in the packet disassembly processing section (7), the packet disassembly section (7a) first separates the audio data into audio data and header information, and the audio data is stored in the fluctuation absorption buffer (7C) in units of packets. The header information is analyzed by the header analysis unit (7b) and the order number is “0”.
” is detected and output as a sound detection signal to the reproduction timing section (7d).

Furthermore, the packet discard monitoring section (7e) checks the sequence of the order numbers, and if there is a gap in the sequence due to packet discard, a packet discard detection signal is output to the reproduction timing control section (7d). The playback timing control section (7d) is a sound detection signal.

Based on the packet discard detection signal, the fluctuation absorption buffer (7
Controls the timing of audio data output from c). The transfer timing of the audio packet is shown in FIG. 61.

In the figure, on the transmitting side, a transmission packet is generated and sent out based on the voice/silence information for each packet from the voice/silence detector (3). At this time, sequential order numbers are assigned to the series of packets 1ζ within the sound section in order starting from "O"'.

The packets received on the receiving side are stored in the fluctuation absorption buffer and then played back. At this time, if ``O'' is detected by checking the sequence number, it is considered as the first packet in the sound section. It is played back after it is known and held in the buffer for Tsec. Packets with order numbers other than "0" are subsequently reproduced continuously (the time T is determined by the configuration of the network connecting the devices, etc.). Furthermore, if a packet is discarded during transmission, this is detected by the discontinuity of the sequence numbers.

[Problem to be solved by the invention]

Since the conventional audio packetization device is configured as described above, it has a function of monitoring and detecting the sequence number ゛0゛° in order to detect the beginning of a sound interval on the receiving side, and detects discarded packets. Two functions are required: checking the sequence of sequence numbers for Also, since the length of the silent section cannot be known, it is susceptible to fluctuations in the silent section.

There were problems such as difficulty in maintaining naturalness between the clear sound part and the sound part during audio playback.

This invention was made to solve the above problems.
The purpose of the present invention is to obtain a voice packet device that can perform the initial detection of a sound section and the detection of whether or not a packet is discarded in the same process on the receiving side, and can also know the length of a silent section. .

[Means to solve the problem]

The audio packetization device according to the present invention adds a sequential order number including the number of packets in a silent section to audio packets of a sound portion to be transmitted, and
This sequence of order numbers is monitored on the receiving side.

[For production]

The audio packetizer of the present invention can simplify packet processing by performing the first detection of a sound part on the receiving side and the detection of packet discard using the same control, and also By knowing this, you can play back silent sections appropriately.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to the drawings. The overall block diagram (FIG. 4) of this embodiment is the same as that of the conventional example, so a description thereof will be omitted.

In FIG. 1, (7) is a packet disassembly processing unit, (7a
) is a packet disassembly unit that separates the packet into audio data and header information, (7b) is a header analysis unit that analyzes the separated header information, and (7c) is a decoding/expanding unit (8) that stores the audio data. A fluctuation absorption buffer for outputting, (7f) a sequence number monitoring unit that monitors the sequence of sequence numbers in the header information, and (7d) a playback unit that controls the output timing of audio data from the fluctuation absorption buffer (7C). This is a timing control section.

Next, the operation will be explained. Packets received from the network side via the line control section (5) are separated into audio data and header information by the packet disassembly section (7a), and the audio data is stored in a fluctuation absorption buffer (7C) in units of packets. The header information is analyzed by a header analysis section (7b) and header processing is performed. Furthermore, the sequence number for each packet is notified from the header analysis unit (7b) to the sequence number monitoring unit (7f), which monitors this and determines whether there is a sound due to discontinuity in the sequence of sequence numbers. Along with detecting intervals and silent intervals.

If a packet is discarded, the period in between is treated as a silent period. Furthermore, the length of the silent section is notified to the reproduction timing control section (7d). Playback timing control section (7d)
Then, the timing of the audio data output from the fluctuation absorption buffer (7c) is controlled based on the voice/silence information and the silent section length from the sequence number monitoring unit (7f). FIG. 2 shows the transfer timing of the voice packets. In the figure,
On the transmitting side, a transmission packet is generated and sent based on the voice/silence information for each packet from the voice/silence detector. At this time, sequential numbers from 0 to 15, including silent periods, are sequentially assigned to the order numbers of the packets to be transmitted.

The sequence number of the packet received on the receiving side is monitored by the sequence number monitoring unit (7f), and since sequence numbers 8 to 11 are missing, the part corresponding to this area is set as a silent area and the secondary Assume that the sound interval begins with the packet with the sequence number "12".Furthermore, the length of the silent interval is determined to be 4 packets.Based on this information, the fluctuation absorption buffer buffers the packet with the sequence number "12". The packet is played after being held within Tsee, and subsequent packets are played continuously.

At this time, the in-buffer hold time T is controlled so that the silent section during audio reproduction has a value close to the silent section length detected by the sequence number monitoring unit (7f). Also.

Even in this case, if a packet is discarded during transmission,
Audio packets are played back using exactly the same procedure.

FIG. 3 shows a frame configuration diagram of a voice packet.

In addition, in the above embodiment, an example was shown in which discontinuity of sequence numbers is used to know the silent interval length on the receiving side, but a packet with information indicating the silent interval length is separately generated on the transmitting side. It is also possible to notify the receiving side.

〔Effect of the invention〕

As described above, according to the present invention, by adding a sequential order number that includes the number of packets in the silent section to the transmitted packet, the processing for the first detection of a sound section and the detection of packet discard on the receiving side are the same. In addition to making it possible to perform this by controlling

Audio packet reproduction hardware and software processing can be simplified, and silent sections can be appropriately reproduced.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a packet disassembly processing section in an audio packetization device according to an embodiment of the present invention. FIG. 2 is a diagram showing the transfer timing of audio packets in one embodiment of the present invention, FIG. 3 is a diagram showing the frame structure of audio packets, and FIG. FIG. 5 is a block diagram of a packetization device, FIG. 5 is a block diagram of a packet disassembly processing section in a conventional audio packetization device, and FIG. 6 is a diagram showing the transfer timing of audio packets in the conventional audio packetization device. In the figure, (1) is an A/D conversion section, (2) is an encoding compression section, (3) is a speech/silence detection section, and (4) is a packet assembly section. (5) is the line system section, (6) is the relay line, (7) is the packet disassembly processing section, (7a) is the packet disassembly section, (7
b) is a header analysis section, (7c) is a fluctuation absorption buffer, (7e) is a packet discard monitoring section, (7f) is a sequence number monitoring section, (7d) is a playback timing control section, (8
) indicates a decoding/expanding section, and (9) indicates a D/A converting section. Note that the same reference numerals in the two figures indicate the same or corresponding parts.

Claims (1)

[Claims] means for encoding and compressing an input audio signal; means for detecting a sound section and a silent section of the audio signal; and means for packetizing data in the sound section and transmitting the packet; and storing received audio packets in a buffer memory. In an audio packetization device having a means for accumulating and a control means for reproducing voice packets that have not been accumulated in the buffer memory after accumulating them for a certain period of time, the number of packets is taken into account when data in a silent section is also packetized in the voice packets. The receiver side monitors the sequence number, learns the sound/silence and silent period length from the discontinuity of the sequence number, and stores it in the buffer memory based on this information. What is claimed is: 1. An audio packetization device that reproduces audio packets that have been processed.