JP2002252644A - Apparatus and method for communicating voice packet - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce a transmission delay in an entire apparatus caused by an unavoidable delay generated by delay elements such as low bit rate encoding, packetizing and transmission lines, etc., and additionally caused by a detecting delay by a DTMF detecting circuit 100 in a voice packet transmission using the low bit rate encoding. SOLUTION: A DTMF detecting part detects DTMF signals in a shorter time than a delay for encoding voice. In a receiving means, packets multiplexed by both voice signals and the DTMF signals are determined as the DTMF signals during a sound interval only when the packets are received the predetermined times or more from a start of the sound interval, otherwise the packets are determined as the voice signals.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice packet communication apparatus for packetizing a voice signal and transmitting the packet, and particularly to a DTMF (Dual Tone Multi) in addition to the voice signal.
i. Frequency (i. Frequency) signal, and a voice packet communication device and a voice packet communication method that eliminate a transmission delay associated with a detection delay of the DTMF signal.

[0002]

2. Description of the Related Art In recent years, with the spread of the Internet and the development of high-efficiency voice coding technology, a voice packet communication technology for coding a telephone voice signal at a low bit rate and transmitting the packet has attracted attention. The signal output from the telephone terminal is not only a voice signal, but also a DTMF for data transmission such as telephone banking and answering machine control.
(Dual Tone Multi Frequency
y; double tone multi-frequency) signal is often used. However, according to a low bit rate coding method of 8 kbit / s or less, which is generally used for voice packet communication and the like, voice signals can be transmitted efficiently without deteriorating the communication quality. Is not possible.

[0003] To solve the above problems, DTM
Japanese Patent Application Laid-Open No. Hei 8-307366 discloses a technique which includes a detector dedicated to the F signal and transmits the speech signal and the DTMF signal with high efficiency in combination with a low bit rate speech encoder.
FIG. 16 is a block diagram showing a configuration of a conventional voice / DTMF signal transmission device disclosed in the above publication. In the figure, reference numeral 100 denotes a DTMF detection circuit for detecting whether or not an input signal is a DTMF signal.
If it is a signal, it outputs an H level DTMF detection signal, and if the input signal is an audio signal, it outputs an L level DTMF detection signal to the switching circuit 101 and the output selection circuit 104. 101
Is the DTMF encoding circuit 102
Alternatively, the switching circuit switches to the voice encoding circuit 103, and upon receiving a DTMF detection signal from the DTMF detection circuit 100, performs the switching operation. 102 is DTMF
A DTMF encoding circuit that encodes the signal outputs the encoded DTMF signal to the output selection circuit 104. 1
Reference numeral 03 denotes an audio encoding circuit for encoding digital audio data, and outputs an encoded audio signal to the output selection circuit 104. Reference numeral 104 denotes a DTMF encoding circuit 10 according to a DTMF detection signal from the DTMF detection circuit 100.
2 is an output selection circuit for selecting one of the output signals of the audio encoding circuit 103.

Next, the operation will be described. First, a voice signal or a DTMF signal is input to a DTMF via an input terminal.
Input to the detection circuit 100 and the switching circuit 101. DTM
The F detection circuit 100 determines whether the input signal is a DTMF signal, and outputs the determination result to the switching circuit 101 and the output selection circuit 104. When the DTMF detection circuit 100 determines that the input signal is a DTMF signal,
An H level DTMF detection signal is output to the switching circuit 101. When receiving the H level DTMF detection signal, the switching circuit 101 outputs an input signal to the DTMF encoding circuit 102. In the DTMF encoding circuit 102, the switching circuit 1
The DTMF signal is input from 01, the number indicated by the DTMF signal is decoded, and output to the output selection circuit 104.

On the other hand, when the DTMF detection circuit 100 determines that the input signal is an audio signal, the L level D
The TMF detection signal is output to the switching circuit 101. This L
Upon receiving the level DTMF detection signal, the switching circuit 101
Outputs an input signal to the speech encoding circuit 103. The audio encoding circuit 103 receives an audio signal from the switching circuit 101, encodes the audio signal at a low bit rate, and performs output encoding on the audio signal.
04. The output selection circuit 104 outputs a DTMF according to the DTMF detection signal from the DTMF detection circuit 100.
The number from the encoding circuit 102 or the speech encoding circuit 10
3 to select and output one of the low bit rate encoded audio signals.

As described above, by providing a dedicated encoding (decoding number) circuit for a DTMF signal, a voice signal and a DTMF signal can be transmitted efficiently. However, if the detection accuracy of the DTMF detection circuit 100 is low, there is a possibility that an audio signal is erroneously determined as a DTMF signal, or conversely, a DTMF signal is erroneously determined as an audio signal. There is a trade-off between the detection accuracy and the detection time, and it is necessary to lengthen the detection time in order to improve the detection accuracy.

For this reason, in practice, a method is used in which a delay circuit corresponding to the time from when a DTMF signal is inputted to when the DTMF detection circuit 100 detects it is inserted immediately before the switching circuit 101. Thus, DTM
An apparatus for delaying a signal by the detection time of the F detection circuit 100 and transmitting a packet of an audio signal and a DTMF signal is disclosed in, for example, JP-A-2000-188611.

[0008]

Since the conventional voice packet communication apparatus is configured as described above, the delay of the low bit rate coding which is inevitably generated in the voice packet transmission using the low bit rate coding. DTMF in addition to delay elements such as packetization delay, delay on the transmission path, etc.
Since the detection delay of the detector 100 is present, there is a problem that the transmission delay of the entire device further increases.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to suppress an increase in transmission delay to a minimum and to reliably transmit a voice signal and a DTMF signal efficiently. An object is to obtain a communication device.

[0010]

A voice packet communication apparatus according to the present invention comprises a voice coding unit for coding an input signal at a low bit rate, and a voice / silence determining unit for determining voice / non-voice of the input signal. And a DTMF detection unit for detecting whether or not the input signal is a DTMF signal and decoding the digit number if the input signal is a DTMF signal, and determining whether the input signal is a sound by the DTMF detection unit and the sound / non-speech determination unit. If the voice signal is determined to be a voice signal, the voice coding unit packetizes the low bit rate coded voice data and outputs it to the packet line, and the voice / non-voice determination unit changes the input signal from voiced to silent. If it is determined that the input signal is sound, the DTMF detection unit and the sound / non-sound determination unit determine that the input signal is sound and DTMF.
A transmitting unit having a packet assembling unit for outputting, to the packet line, a packet in which the audio data is multiplexed with the low bit rate encoded audio data and the digit number of the DTMF signal; When a packet in which low-bit-rate coded audio data and a digit number of a DTMF signal are multiplexed and received for a predetermined number of times consecutively from the first packet of the voiced section of the voiced section of the packet, the voiced section is determined to be a DTMF signal. Otherwise, a received signal type determination unit that determines an audio signal, a buffer that stores received packets until a result of the determination by the received signal type determination unit is obtained, and a voiced interval is determined by the received signal type determination unit. An audio decoding unit that decodes low bit rate encoded audio data included in a received packet determined to be an audio signal , DT based on the digit number of the DTMF signals voiced interval by receiving the signal type determination unit is included in the received packet is determined to be a DTMF signal
And a receiving unit having a DTMF signal generating unit for generating an MF signal.

[0011] The voice packet communication device according to the present invention comprises:
The DTMF detection unit detects whether or not the input signal is a DTMF signal in a time equal to or less than the encoding delay of the voice encoding unit.

[0012] A voice packet communication device according to the present invention comprises:
When the received signal type determination unit receives the voiced end notification packet and does not determine the signal type of the voiced section, the signal type is determined to be an audio signal.

A voice packet communication device according to the present invention comprises:
Even if the first packet is received from the packet line after the reception signal type determination unit turns into a sound and a predetermined time or more elapses, if the signal type of the sound period is not determined, the signal type is determined. It is determined as an audio signal.

[0014] The voice packet communication device according to the present invention comprises:
Whether the packet assembler is a packet containing only the low bit rate encoded voice data, a packet containing only the decoding result of the digit number of the DTMF signal, or the low bit rate encoded voice data and the DTMF signal. A different packet header is added depending on whether the result of decoding the digit number is a multiplexed packet.

A voice packet communication device according to the present invention comprises:
When the packet including the digit number of the DTMF signal is lost, the DTMF signal generation unit generates the DTMF signal based on the result of decoding the digit number included in the packet received immediately before the lost packet.

A voice packet communication device according to the present invention comprises:
The packet assembler multiplexes the low bit rate coded voice data and the digit number of the DTMF signal by the voice coder continuously for a predetermined number of times including the first output packet after the input signal has sound. After the output of the packet, a packet containing only the digit number of the DTMF signal is output until the output of the voiced end notification packet.

The voice packet communication device according to the present invention comprises:
The receiving means includes a reception signal level calculation unit for calculating the signal level of the signal output from the audio decoding unit, and the DTMF signal generation unit controls the DTMF signal by controlling the signal level equal to the signal level calculated by the reception signal level calculation unit. Output.

The voice packet communication device according to the present invention comprises:
The reception signal level calculation unit calculates the mean square of the signal levels of the signals output from the speech decoding unit, and uses the mean square value as the signal level.

The voice packet communication device according to the present invention comprises:
The reception signal level calculator calculates an average of the absolute values of the signal levels of the signals output from the audio decoder, and uses the average of the absolute values as the signal level.

The voice packet communication device according to the present invention comprises:
The reception signal level calculation unit calculates the maximum value of the signal level of the signal output from the audio decoding unit, and uses this maximum value as the signal level.

[0021] The voice packet communication device according to the present invention comprises:
A gain information extracting unit that extracts gain information included in the audio data of the packet received by the low bit rate coded low bit rate, and calculates a signal level of a signal output by the audio decoding unit based on the gain information; The signal generator controls the signal level to be equal to the signal level calculated by the gain information extractor and outputs a DTMF signal.

[0022] The voice packet communication device according to the present invention comprises:
The transmitting means includes a transmission signal level calculator for calculating the signal level of the input signal, and the packet assembler multiplexes the digit number of the DTMF signal and the signal level calculated by the transmission signal level calculator, or a low bit rate packet. A packet in which the coded voice data, the digit number of the DTMF signal and the signal level calculated by the transmission signal level calculation unit are multiplexed is output, and the signal level equal to the signal level extracted from the packet received from the transmission unit by the DTMF signal generation unit And outputs a DTMF signal.

[0023] The voice packet communication device according to the present invention comprises:
The transmission signal level calculator calculates the mean square of the signal levels of the input signals, and uses the mean square value as the signal level.

The voice packet communication device according to the present invention comprises:
The transmission signal level calculator calculates the average of the absolute values of the signal levels of the input signals, and uses the average of the absolute values as the signal level.

The voice packet communication device according to the present invention comprises:
The transmission signal level calculator calculates the maximum value of the signal level of the input signal, and uses this maximum value as the signal level.

The voice packet communication device according to the present invention comprises:
If the result of the determination as to whether or not the DTMF detector is a DTMF signal determines that the input signal is not a DTMF signal until a predetermined time elapses after the input signal has sound, then the DTMF signal is output until no sound is generated. It is not determined.

The voice packet communication device according to the present invention comprises:
When the packet assembler outputs a packet including only the low bit rate coded audio data after the input signal becomes voiced, the low bit rate coded voice data is output until a voiced end notification packet is output thereafter. It outputs only packets.

The voice packet communication method according to the present invention comprises:
In a voice packet communication method for generating a packet based on an input signal from a terminal device and outputting the packet to a packet line, and receiving the packet from the packet line, a voice coding step of coding the input signal at a low bit rate; If the signal is a voice signal and a voice signal, the voice data coded at the low bit rate in the voice coding step is packetized and output to a packet line. When the input signal changes from voice to silence, a voice termination notification packet Is output to the packet line, and if the input signal is a sound and a DTMF signal, a packet obtained by multiplexing the low bit rate encoded audio data and the digit number of the DTMF signal in the audio encoding step is output to the packet line. A transmitting step having a packet assembling step, and a voice section of a packet received from the packet line. When a packet in which low-bit-rate coded voice data and a DTMF signal digit number are multiplexed from a packet is continuously received a predetermined number of times, the voiced section is determined to be a DTMF signal. A received signal type determining step of determining a signal; and a voice decoding step of decoding low bit rate coded voice data included in the received packet whose voiced section is determined to be a voice signal in the received signal type determining step. Generating a DTMF signal based on the digit number of the DTMF signal included in the received packet whose sound period is determined to be a DTMF signal in the received signal type determination step.
And a receiving step having a TMF signal generating step.

The voice packet communication method according to the present invention comprises:
It detects whether or not the input signal is a DTMF signal in a time shorter than the coding time in the voice coding step.

[0030]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below. Embodiment 1 FIG. FIG. 1 is a block diagram showing a configuration of a voice packet communication device according to Embodiment 1 of the present invention.
(A) shows the configuration of the transmitting means, and (b) shows the configuration of the receiving means. In the figure, 1 is a DTM from an input signal.
A DTMF detector that detects the F signal and decodes the digit number if it is a DTMF signal, 2 is a voice encoder that encodes the input signal at a low bit rate, and 3 is a voiced / non-voiced input signal. 4 is a packet assembling section that packetizes voice signals and DTMF signals, 4a is a payload multiplexing section that assembles a payload portion of a packet with the components of the packet assembling section 4, and 4b is a packet assembling section 4. It is a header adding unit that adds a packet header as a component. The DTMF detection unit 1, the voice coding unit 2, the voice / non-voice determination unit 3, and the packet assembling unit 4 constitute a transmission unit.

Reference numeral 5 denotes a buffer unit (buffer) for storing a packet input from the packet line side, 6 a received signal type determination unit for monitoring the type of the input packet to determine the type of the received signal, and 7 a buffer unit. A packet readout controller for reading out the packet stored in 5; a packet decomposer 8 for extracting a payload portion of the packet; 9 a voice decoder for decoding low bit rate encoded voice data; 10 a DTMF signal Is a DTMF signal generation unit for generating a noise, 11 is a silence generation unit for generating pseudo background noise, and 12 is an output selection unit. The buffer unit 5, the received signal type determining unit 6, the packet reading control unit 7, the packet decomposing unit 8, the audio decoding unit 9, the DTMF signal generating unit 10, the silence generating unit 11, and the output selecting unit 12 constitute a receiving unit. You.

Next, the operation will be described. First, the operation on the transmitting side of the voice packet communication device according to the first embodiment will be described. A voice packet communication device according to the present invention packetizes an input signal from a terminal device and sends the packet to a packet line, receives a packet from the packet line, outputs the decoded content to the terminal device, and performs communication. First, as shown in FIG. 1A, a signal input from a terminal device (not shown) is input to a DTMF detection unit 1, a voice coding unit 2, and a sound / non-speech determination unit 3. The voice coding unit 2 outputs a low bit rate coded signal obtained by coding the low bit rate signal to the payload multiplexing unit 4a regardless of whether the input signal is a voice signal or a DTMF signal. For the sake of convenience in the following description, the low bit rate encoding method used in the audio encoding unit 2 is described in ITU-T Recommendation G. 8 specified in 729
kbit / sCS-ACELP (Conjugate
Structure Algibraic Code-
Excited Linear Prediction
n) method.

The DTMF detection section 1 outputs the DTMF
It determines whether or not the signal is a signal and decodes the number, outputs the result of determination to the payload multiplexing unit 4a, and when the input signal is a DTMF signal, outputs the decoded digit number (hereinafter referred to as the DTMF number) to the payload. Output to the multiplexing unit 4a. Here, the DTMF detection unit 1 sacrifice a certain degree of detection accuracy and suppresses the detection time. Specifically, the detection time is set to be equal to or less than the encoding delay of the audio encoding unit 2, and in the 8 kbit / s CS-ACELP system, the encoding delay is 15 ms. Is 15
ms or less. On the other hand, the sound / silence determination unit 3
Then, a sound / non-sound of the input signal is determined, and the determination result is output to the payload multiplexing unit 4a.

The payload multiplexing unit 4a determines that the input signal from the terminal device is a voice and a voice signal based on the determination results input from the DTMF detection unit 1 and the voice / non-voice determination unit 3 (DTMF signal If not, the low-bit-rate encoded audio signal input from the audio encoding unit 2 is output to the header adding unit 4b. When the input signal from the terminal device is a sound and a DTMF signal, the low bit rate coded voice signal input from the voice coding unit 2 and the DTMF number input from the DTMF detection unit 1 are used. These are input, multiplexed, and output to the header adding unit 4b.
If the input signal is not a sound, the payload multiplexing unit 4a does not output anything to the header adding unit 4b.

The header adding section 4b adds a packet header to the signal input from the payload multiplexing section 4a based on the determination result from the DTMF detecting section 1, and outputs the signal to a packet line (not shown). In addition, when the input signal changes from voiced to voiceless, a voiced end notification packet indicating the end of the voiced section is output. One packet is output from the packet assembling unit 4 including the payload multiplexing unit 4a and the header adding unit 4b every 10 ms while the input signal from the terminal device is a sound. This is C
S-ACELP coded speech coded frames are 10
ms.

Here, the operation of the packet assembling section 4 composed of the payload multiplexing section 4a and the header adding section 4b will be described. 2A and 2B are diagrams showing a format of a packet output by the packet assembling unit. FIG. 2A shows a packet of only an audio signal, FIG. 2B shows a packet in which an audio signal and a DTMF signal are multiplexed, and FIG. ) Indicates a voiced end notification packet. As shown in FIG. 2, the packet header is composed of a sequence number (SN) and a packet type, and the payload is composed of only low bit rate encoded audio data and a DTMF number, or only low bit rate encoded audio data. Sequence number (SN) is 1
This is a counter value that is incremented by one each time a packet is output. The packet type is a packet indicating the contents of the payload.
As shown in FIG. 2A, when the payload is low bit rate audio encoded data and the packet type is 1,
As shown in (b), the payload is data obtained by multiplexing low bit rate encoded voice data and a DTMF number.
When the packet type is 2, no payload exists as shown in FIG. 2 (c), indicating that the voiced section ends with this packet.

Next, the operation on the receiving side of the voice packet communication device according to the first embodiment will be described. First, packets received from the packet line side are stored in the buffer unit 5 and input to the received signal type determination unit 6. From the beginning of the voiced section of the received packet, the received signal type determination unit 6 continuously outputs, for example, three low bit rate encoded speech data and D
When a packet multiplexed with the TMF signal is received, the voiced section is determined to be a DTMF signal, otherwise, it is determined to be a voice signal, and the determination result is output to the packet read control unit 7. I do.

Therefore, when a packet in which three low bit rate coded voice data and a DTMF signal are multiplexed consecutively from the beginning of the voiced section is received, a decision result is output at the stage when the third packet is received. However, if the first packet of the voiced section is the low bit rate coded voice data, the voiced section is immediately determined as a voice signal, so that the determination result can be output at the same time as the reception of the top packet.

When the judgment of the received signal type judging section 6 is completed, the packet reading control section 7 reads out the received packet stored in the buffer section 5 according to the judgment result, outputs the read packet to the packet decomposing section 8, and outputs the packet. A control signal for instructing the selection unit 12 to output any of the output signals of the speech decoding unit 9, the DTMF signal generation unit 10, and the silence generation unit 11 to the terminal device side is output.

The packet disassembly unit 8 extracts the payload of the packet input from the packet readout control unit 7, and outputs the low bit rate encoded audio data to the audio decoding unit 9.
The MF number is output to the DTMF signal generator. The audio decoder 9 decodes the low bit rate encoded audio data input from the packet decomposer 8 and outputs it to the output selector 12. Further, the DTMF signal generator 10 outputs a DTMF signal corresponding to the DTMF number input from the packet decomposer 8. The signal level of the DTMF signal output from the DTMF signal generator 10 is a standard value,
For example, it is fixed at -15 dBm. On the other hand, the silence generating section 11 outputs a silence signal to the output selecting section 12. The output selection unit 12 responds to a control signal input from the packet read control unit 7 by a voice decoding unit 9, a DTMF signal generation unit 10,
And the output signal of any one of the silent generator 11 is selected and output to the terminal device side.

Here, the operation of the packet read control unit 7 will be described in detail. FIGS. 3 to 8 show the relationship between the packet sequence received by the receiving unit of the voice packet communication device according to the first embodiment, the output of the packet readout control unit (input of the packet decomposing unit), and the control signal input to the output selection unit. FIG. The operation of the packet read control unit 7 will be described with reference to these figures. As shown in FIG. 3, when the first packet of the voiced section of the received packet sequence is low bit rate coded voice data, the received signal type determination unit 6 determines that the voiced section is voice, and Is output to the packet read control unit 7. Upon receiving this determination result, the packet read control unit 7 immediately reads the packet from the buffer unit 5 and outputs it to the packet decomposing unit 8, and selects the output of the audio decoding unit 9 to the output selecting unit 12. (The input value of the selection unit input is selected as the audio level as shown in FIG. 3). Thereafter, the operation of reading packets from the buffer unit 5 and outputting the packets to the packet disassembly unit 8 every packet reception cycle (10 ms) while maintaining the control signal state for the output selection unit 12 is continued. At the time when the packet received just before the voiced end notification packet (END flag) of the received packet sequence has been output to the packet decomposing unit 8, the packet reading control unit 7
The control signal is changed so that the output of the silence generation unit 11 is selected and output for No. 2 (the input value of the selection unit input is changed to a silence level as shown in FIG. 3).

When the head packet of a voiced section is low-bit-rate coded voice data, the received signal type determination section 6 always regards the voiced section as voice.
As a result, as shown in FIG. 4, a packet in which the low bit rate coded voice data and the DTMF number are multiplexed in the middle of the received packet sequence is received until the voiced end notification packet is received. Even in the case, the output selection unit 12
Is not changed (the input value of the selection unit input is selected as the audio level).

As shown in FIG. 5, when the first packet of the sound section of the received packet sequence is a packet in which low bit rate encoded voice data and a DTMF number are multiplexed, for example, up to three packets are received. The packet read control unit 7 does not read packets from the buffer unit 5. When a packet in which low-bit-rate encoded voice data and a DTMF number are multiplexed for three consecutive packets is received, the received signal type determination unit 6 determines that the sound period is DT.
It regards it as an MF signal and outputs a signal indicating that to the packet readout control unit 7.

Upon receiving the determination result from the received signal type determination unit 6, the packet read control unit 7 starts reading the received packet at that time and outputs it to the packet disassembly unit 8. Then, a control signal is output so as to select the output of the DTMF signal generator 10 (the input value of the input of the selector is selected to the level of DTMF as shown in FIG. 5). Thereafter, the operation of reading one packet from the buffer unit 5 and outputting it to the packet decomposing unit 8 every packet reception cycle (10 ms) is continued while maintaining the control signal state for the output selection unit 12. Voice end notification packet of received packet sequence (END flag)
At the point in time when the packet received immediately before is output to the packet decomposing unit 8, the output selecting unit 12 is
The control signal is changed so as to select and output the output No. 1 (the input value of the selection unit input is changed to a silent level as shown in FIG. 5).

The first packet of the voiced section of the received packet sequence is composed of low bit rate encoded voice data and DTMF.
As shown in FIG. 6, if a packet including only low-bit-rate coded audio data is received within three packets even if the packet is a packet in which the number is multiplexed, the received signal type determination unit 6 Is regarded as an audio signal, and a signal indicating this is output to the packet readout control unit 7.

Upon receiving the determination result from the received signal type determination unit 6, the packet read control unit 7 starts reading the received packet at that time and outputs it to the packet disassembly unit 8. And outputs a control signal so as to select the output of the audio decoding unit 9 (the input value of the selection unit input is selected as the audio level as shown in FIG. 6). Then, the operation of reading one packet from the buffer unit 5 every packet reception cycle (10 ms) and outputting it to the packet decomposing unit 8 is continued while maintaining the state of the control signal for the output selection unit 12. When the output of the packet received immediately before the voiced end notification packet (END flag) of the received packet sequence to the packet decomposing unit 8 is completed, the output of the silence generating unit 11 is selected by the output selecting unit 12. The control signal is changed so as to be output (the input value of the selection unit input is changed to a silent level as shown in FIG. 6).

Further, the first packet of the sound period of the received packet sequence is composed of low bit rate coded voice data and DTM.
Even if the F-number is multiplexed with a packet, if a voiced end notification packet (END flag) is received within three packets, the received signal type determination unit 6 determines that the voiced section is a voice signal. Then, a signal indicating that fact is output to the packet read control unit 7.

Upon receiving the determination result from the received signal type determination unit 6, the packet read control unit 7 starts reading the received packet at that time and outputs it to the packet disassembly unit 8. And outputs a control signal so as to select the output of the audio decoding unit 9 (the input value of the selection unit input is selected as the audio level as shown in FIG. 7). Thereafter, the operation of reading one packet from the buffer unit 5 and outputting it to the packet decomposing unit 8 every packet reception cycle (10 ms) is continued while maintaining the control signal state for the output selection unit 12. When the packet received immediately before the voiced end notification packet (END flag) of the received packet sequence has been output to the packet decomposing unit 8,
The control signal is changed to the output selection unit 12 so as to select and output the output of the silence generation unit 11 (the input value of the selection unit input is changed to a silence level as shown in FIG. 7).

Further, the first packet of the sound period of the received packet sequence is composed of low bit rate encoded voice data and DTM.
F-number and multiplexed packet, for example, 30 ms
If the third packet is not received even after the predetermined time elapses, the received signal type determination unit 6 considers that the voiced section is voice and that the voiced end notification packet has been lost on the transmission path. Then, a signal indicating this is output to the packet readout control unit 7 (as shown in FIG. 8, the input value of the selection unit input after 30 ms is selected as the audio level).

Upon receiving the determination result from the received signal type determination unit 6, the packet read control unit 7 reads the packet from the buffer unit 5 at that time and outputs the packet to the packet disassembly unit 8, and outputs the packet to the output selection unit 12. On the other hand, it outputs a control signal so as to select the output of the audio decoding unit 9. Thereafter, the operation of reading packets from the buffer unit 5 and outputting the packets to the packet disassembly unit 8 every packet reception cycle (10 ms) while maintaining the control signal state for the output selection unit 12 is continued. When the output of the packet received immediately before the voiced end notification packet (END flag) of the received packet sequence to the packet decomposing unit 8 is completed, the output selecting unit 1
The control signal is changed so that the output of the silence generating unit 11 is selected and output for No. 2 (the input value of the selection unit input is changed to the level of silence as shown in FIG. 8).

FIG. 9 shows the output of the packet reading control unit (input of the packet decomposing unit), the input of the DTMF signal generation unit, and the output of the packet reading control unit when the packet sequence received by the receiving unit of the voice packet communication apparatus according to the first embodiment has a loss. FIG. 4 is a diagram illustrating a relationship between the control signal and a control signal input to an output selection unit. In the figure, the received packet sequence, the output of the packet readout control unit 7 to the packet decomposing unit 8, and the output selection unit 1
2 and a DTMF signal generator 10
2 also shows a signal output by the packet decomposing unit 8. As shown in FIG.
When a received packet is lost, the output to the MF signal generator 10 maintains the DTMF number of the packet received immediately before the packet. By doing so, it is possible to output a normal DTMF signal even when a received packet is lost.

As described above, according to the first embodiment, even if the input signal is determined to be a DTMF signal by the transmitting means, the audio data subjected to the low bit rate encoding is converted to the DTMF signal.
When a packet in which low-bit-rate coded voice data and a DTMF number are multiplexed is received continuously for a predetermined number of times (for example, three packets), the sound section of the received packet is transmitted to the DTMF. Operate as if it were a signal. Therefore, DT in the transmitting means
Even if the detection accuracy of the MF detection unit 1 is poor and the voice signal is erroneously determined to be a DTMF signal, if the erroneous determination state does not continue for a predetermined number of times (for example, three packets, that is, for 30 ms), the DTMF signal is received by the receiving means. Because it is not considered
Even if the detection time is shortened by sacrificing the detection accuracy of the DTMF detection unit 1, the influence of the erroneous determination can be finally eliminated.

For a voice signal which is not determined to be a DTMF signal at all, the receiving unit immediately decomposes the received packet and decodes it in the voice decoding unit 9. Therefore, it is not necessary to insert a delay to improve the detection accuracy of the DTMF signal in most of the sound section, and the transmission of the voice signal and the DTMF signal is reliably performed while minimizing the increase in the transmission delay of the voice signal. It is possible to obtain a voice packet communication device capable of performing the above.

Embodiment 2 In the first embodiment, an example has been described in which a packet determined by the packet assembling unit 4 to be a DTMF signal always outputs a packet obtained by multiplexing low-bit-rate encoded voice data and a decrypted DTMF number. In the second embodiment, when the DTMF detection state lasts for 30 ms or more (this is a time corresponding to three packets), only the DTMF number is packetized.

FIG. 10 is a diagram showing a packet format handled by the voice packet communication device according to the second embodiment of the present invention. As shown in the figure, the DTMF detection state is 30 m
s (this is the time equivalent to three packets)
The packet assembler 4 forms a packet from a packet header including a sequence number (SN) and a packet type, and a payload storing a DTMF number. here,
Packet type = 3 is assigned to the DTMF number. In addition,
The same components as those in FIG. 2 are denoted by the same reference numerals, and redundant description will be omitted.

FIG. 11 shows a received packet sequence when the receiving means of the voice packet communication apparatus according to the second embodiment detects a DTMF signal continuously for 30 ms or more (for three packets), and the output of the packet read control unit (FIG. 11). FIG. 9 is a diagram illustrating a relationship between a control signal input to an output selection unit and an input of a packet decomposition unit. As shown in FIG. 11, a packet having only the DTMF number is output from the fourth packet of the received packet sequence (from the fourth packet of the received packet sequence, the input value of the selector input is selected to the DTMF level).

Further, in the above embodiment, when a DTMF signal is detected continuously for 30 ms or more (for three packets), the output of the packet readout control unit 7 to the packet disassembly unit 8 and the control signal to the output selection unit 12 are transmitted. Although the output is also shown, if the packet read control unit 7 treats the packet having only the DTMF number in exactly the same way as the packet in which the low bit rate coded voice data and the DTMF number are multiplexed, the DTMF can be performed without any problem. A signal can also be generated.

As described above, according to the second embodiment, when the state determined to be a DTMF signal continues for a predetermined time (for example, 30 ms corresponding to three packets), low bit rate encoded voice data is multiplexed. Don't, DTM
Since only the F number is packetized and output, the packet size is reduced accordingly, and transmission efficiency can be improved.

Further, according to the second embodiment and the first embodiment, a packet containing only low-bit-rate encoded voice data or a packet containing only the decoding result of the digit number of the DTMF signal is used. A different packet header is added depending on whether the packet is a multiplexed packet of voice data encoded at a low bit rate and the decoding result of the digit number of the DTMF signal. Can be deciphered.

Embodiment 3 In the first embodiment, D
Although the example in which the output signal level of the TMF signal generator 10 is set to a fixed level has been described, in the third embodiment, the level of the DTMF signal generated in the receiving means is changed according to the level of the DTMF signal input in the transmitting means. To change it.

FIG. 12 is a diagram showing a configuration of a voice packet communication device according to the third embodiment of the present invention, wherein (a) shows the configuration of the transmitting means and (b) shows the configuration of the receiving means.
In the figure, reference numeral 10a denotes a DTMF signal generator that generates a DTMF signal at a signal level determined based on the signal level calculated by the level calculator 13, and 13 denotes a level calculator that calculates the signal level of the signal output by the audio decoder 9. (Reception signal level calculation unit). The same components as those in FIG. 1 are denoted by the same reference numerals, and duplicate description will be omitted.

Next, the operation will be described. The operation of the transmitting unit is the same as that of the first embodiment, and the buffer unit 5, the received signal type determining unit 6, the packet reading control unit 7, the packet decomposing unit 8, the audio decoding unit 9, the silence generating unit 11, The output selector 12 is also the same as in the first embodiment.
Performs the same operation as the voice packet communication device shown in FIG. Here, different points from the first embodiment will be described. First, the PCM (Pulse Pulse) output from the audio decoding unit 9 is output.
The level calculator 13 calculates the level of a code modulation (Code Modulation) signal, and outputs the calculation result to the DTMF signal generator 10a. The signal level can be calculated by the level calculator 13 by, for example, finding the mean square of the signal level and using the mean square value as the signal level.

When receiving the signal level calculated by the level calculator 13, the DTMF signal generator 10 a outputs a DTMF signal corresponding to the DTMF number input from the packet decomposer 8 at the signal level input from the level calculator 13. I do.

As described above, according to the third embodiment, the receiving means includes the level calculating section 13 for calculating the signal level of the signal output from the audio decoding section 9, and the DTMF signal generating section 10a performs the level calculating. Since the DTMF signal is output by controlling the signal level to be equal to the signal level calculated by the unit 13, the transmitting unit outputs the DTMF signal input from the terminal device (not shown).
The receiving means can output the DTMF signal equal to the signal level of the F signal, and the DTMF signal can be transmitted with higher accuracy.

In the third embodiment, the signal level of the PCM signal is obtained by the mean square in the level calculating section 13. However, the absolute value or the maximum value of the signal level of the PCM signal may be substituted. Similar effects can be obtained. In addition, the invention of the present application is not limited to the above, provided that an appropriate value can be calculated as the signal level of the PCM signal.
It is not limited to the above method.

Embodiment 4 The fourth embodiment shows another configuration example in which the transmitting means changes the signal level of the DTMF signal generated by the receiving means according to the signal level of the DTMF signal input from the terminal device.

FIG. 13 is a diagram showing a configuration of a voice packet communication device according to a fourth embodiment of the present invention. FIG. 13 (a) shows the configuration of a transmitting unit, and FIG. 13 (b) shows the configuration of a receiving unit.
In the figure, reference numeral 10b denotes a DTMF signal generator for generating a DTMF signal equal to the signal level of the DTMF signal inputted by the transmitting means based on the gain information extracted by the gain information extractor, and 14 denotes a gain of the low bit rate encoded voice data. It is a gain information extraction unit that extracts information. FIG.
The same components as those described above are denoted by the same reference numerals, and redundant description will be omitted.

Next, the operation will be described. The operation of the transmitting unit is the same as that of the first embodiment, and the buffer unit 5, the received signal type determining unit 6, the packet reading control unit 7, the packet decomposing unit 8, the audio decoding unit 9, the silence generating unit 11, The output selector 12 is also the same as in the first embodiment.
Performs the same operation as the voice packet communication device shown in FIG. Here, different points from the first embodiment will be described. First, the gain information extraction unit 14 determines whether the packet decomposition unit 8
Out of the low bit rate encoded audio data output by
Extract the gain information part. The low bit rate encoded audio data includes gain information. For example, when the signal level is −40 dBm or less, the gain information is 0, and −40 to −40.
It is 1 at 35 dBm and 2 at -35 to -30 dBm.
And -3 at -30 to -25 dBm, and -25 to -25 dBm.
It is 4 at −20 dBm, 5 at −20 to −15 dBm, 6 at −15 to −10 dBm, and −10.
The code is substantially determined according to the range of the signal level, such as 7 when the value is equal to or larger than dBm. Therefore,
In the case of this example, the level of the input signal can be known with an error within 5 dB from the gain information. Therefore, the gain information extracting unit 14 determines a signal level based on the gain information, and determines the determined level as DTMF.
The signal is output to the signal generator 10b.

In the DTMF signal generator 10b, the DTMF corresponding to the DTMF number input from the packet
The signal is output at the signal level input from the gain information extraction unit 14.

As described above, according to the fourth embodiment, the gain information included in the audio data at the low bit rate of the received packet is extracted, and the audio decoding unit 9 outputs the gain information based on the gain information. A gain information extracting unit 14 for calculating a signal level of the signal to be output, and the DTMF signal generating unit 10b outputs a DTMF signal by controlling the signal level to be equal to the signal level calculated by the gain information extracting unit 14. The receiving unit can output a DTMF signal that is substantially equal to the signal level of the DTMF signal input from the device, and more accurate transmission of the DTMF signal becomes possible.

Embodiment 5 FIG. This fifth embodiment shows a configuration example different from the third and fourth embodiments in which the transmitting means changes the signal level of the DTMF signal generated by the receiving means according to the signal level of the DTMF signal input from the terminal device. Things.

FIG. 14 is a diagram showing a configuration of a voice packet communication device according to the fifth embodiment of the present invention. FIG. 14 (a) shows the configuration of the transmitting means, and FIG. 14 (b) shows the configuration of the receiving means.
In the figure, reference numeral 4A denotes a packet assembling section having a payload multiplexing section 4c, and 4c denotes a payload multiplexing section (packet assembling section) which is a component of the packet assembling section 4A. Calculation unit 1
5, or a packet obtained by multiplexing the low bit-rate coded audio data with the digit number of the DTMF signal and the signal level calculated by the level calculator 15. 8a is a packet decomposing unit that extracts a signal level from the received packet by the level calculating unit 15 multiplexed by the payload multiplexing unit 4c and outputs the signal level to the DTMF signal generating unit 10c. 10c is a signal calculated by the level calculating unit 15 received from the transmitting unit. A DTMF signal generator for generating a DTMF signal equal to the level, 15
Denotes a level calculator (transmission signal level calculator) that calculates the signal level of an input signal from a terminal device (not shown).
The same components as those in FIG. 1 are denoted by the same reference numerals, and duplicate description will be omitted.

Next, the operation will be described. DTMF detection unit 1, voice coding unit 2, voiced / silence determination unit 3, header addition unit 4b in the transmission unit, and buffer unit 5, received signal type determination unit 6, packet read control unit 7, voice in the reception unit The decoding unit 9, the silence generation unit 11, and the output selection unit 12 perform the same operation as the voice packet communication device described in the first embodiment. Here, different points from the first embodiment will be described.

First, the level calculation section 15 calculates the signal level of the input signal from the terminal device by means of the mean square, codes this, and outputs it to the payload multiplexing section 4c. At the time of this coding, a method of outputting the code x in the case of -xdBm is used. For example, if the signal level is -10 dBm, a code of 10 is output. In the case of a telephone voice signal, the dynamic range of an input signal is in the range of about -65 dBm to 0 dBm. Therefore, using such coding enables coding with 7-bit information.

The operation of the payload multiplexing unit 4c is almost the same as that of the voice packet communication device shown in FIG.
When multiplexing the low bit rate encoded voice data and the DTMF number, or when outputting only the DTMF number, the level code of the DTMF signal is further multiplexed. FIG. 15 is a diagram showing a packet format of the voice packet communication device according to the fifth embodiment. In the receiving means, the DTMF signal level code (DT
MF level) is extracted by the packet decomposing unit 8a,
The DTMF signal generation unit 10c is notified. As a result, the DTMF signal generator 10c outputs a DTMF signal having a signal level according to the DTMF signal level code.

As described above, according to the fifth embodiment, the transmitting means includes the level calculating section 15 for calculating the signal level of the input signal, and the packet assembling section 4A calculates the digit number of the DTMF signal and the transmission signal level. A packet multiplexed with the signal level calculated by the transmission unit or a packet obtained by multiplexing the audio data subjected to low bit rate encoding with the digit number of the DTMF signal and the signal level calculated by the transmission signal level calculation unit is output. Although the packet size when the number is transmitted increases, the signal level once encoded and decoded at a low bit rate is not always exactly the same as the original signal level. Considering the fact that the gain information in the middle is level information with a rather coarse accuracy, the gain information shown in the third and fourth embodiments is considered. Compared to voice packet communication device, receiving means it is possible to generate a more accurately reproduce DTMF signal a signal level of the input DTMF signal transmitting means.

In the fifth embodiment, the signal level of the PCM signal is obtained by the mean square in the level calculation section 15. However, even if the absolute value or the maximum value of the signal level of the PCM signal is substituted, The same effect can be obtained. In addition, the invention of the present application is not limited to the above, provided that an appropriate value can be calculated as the signal level of the PCM signal.
It is not limited to the above method.

According to the voice packet communication device described in the first to fifth embodiments, the DTMF detection unit 1
As a result, there is a possibility that a packet in which low bit rate coded voice data and a DTMF number are multiplexed is output in the middle of a voiced section as shown in FIG. However,
The DTMF detection unit 1 determines that the DTMF is detected within 30 ms after the input signal on the terminal device becomes sound.
If it is determined that the signal is not the F signal, the operation is performed so that the signal is not determined to be a DTMF signal thereafter until no sound is generated. Can be prevented from outputting a multiplexed packet. This makes it possible to prevent an increase in the packet size.

Further, in the first to fifth embodiments,
When the packet assembling unit 4 or 4A outputs a packet whose payload is only low bit rate encoded audio data after the input signal becomes voiced, the packet assembling unit 4 and 4A output the packet until the voiced end notification packet is output. Can output only packets of low bit rate encoded audio data, the same effect as in the first embodiment can be obtained.

Further, in the first to fifth embodiments,
The received signal type determination unit 6 determines that the low bit rate encoded audio data and DT
When a packet in which the MF number is multiplexed is received, this voiced section is determined to be a DTMF signal. However, the predetermined number of continuous receptions may be different depending on the detection accuracy of the DTMF detector 1. If this detection accuracy is poor,
A continuation of four packets may be used as a determination condition that the signal is a DTMF signal. Conversely, if detection accuracy is good, two packets may be used continuously.

[0081]

As described above, according to the present invention, on the transmitting side, an input signal from a terminal device is coded at a low bit rate, and when the input signal is a voice signal and a voice signal, it is coded at a low bit rate. The voice data is packetized and output to the packet line. When the input signal changes from voiced to silent, a voiced end notification packet is output to the packeted line. If the input signal is voiced and the signal is a DTMF signal, the low bit rate code is output. A packet in which the coded voice data and the digit number of the DTMF signal are multiplexed is output to the packet line, and the receiving side receives the low bit rate coded voice data and DT from the first packet of the voiced section of the packet received from the packet line.
When a packet in which the digit number of the MF signal is multiplexed with the digit number is continuously received a predetermined number of times, the voiced section is determined to be a DTMF signal, otherwise, it is determined to be a voice signal, and the voiced section is a voice signal. If it is determined that there is, the low-bit-rate coded audio data included in the received packet is decoded, and if the voiced section is determined to be a DTMF signal, it is determined based on the digit number of the DTMF signal included in the received packet. DTMF signal is generated, the detection accuracy of the DTMF detector in the transmitting means is low,
Even if an erroneous determination is made as a TMF signal, if the erroneous determination state does not continue for a predetermined number of times, the receiving means does not regard the signal as a DTMF signal. Therefore, there is an effect that the influence of erroneous determination of the DTMF signal due to the above can be reduced.

According to the present invention, the transmission means detects whether or not the input signal is a DTMF signal in a time shorter than the voice coding delay, so that the influence of the detection delay of the DTMF detector on the transmission delay of the entire apparatus. This has the effect that it can be reduced.

According to the present invention, if the signal type of the voiced section is not determined when the voiced end notification packet is received, the signal type is determined to be an audio signal. Has the effect that it is not necessary to insert a delay in order to improve the DTMF signal detection accuracy, and it is possible to reliably transmit the voice signal and the DTMF signal while minimizing the increase in the transmission delay of the voice signal. .

According to the present invention, even if the first packet has been received from the packet line after the sound has been output and a predetermined time or more has elapsed, and the signal type determination result of the sounded section has not been output, the signal is output. Since the signal type is determined to be a voice signal, the DTMF is suppressed while minimizing the increase in the transmission delay of the voice signal.
There is an effect that the influence of erroneous determination of the DTMF signal by the detection unit can be reduced.

According to the present invention, whether the packet contains only low-bit-rate encoded audio data,
A packet containing only the decoding result of the digit number of the F signal, or low bit rate encoded voice data and D
Since a different packet header is added depending on whether the decoding result of the digit number of the TMF signal is a multiplexed packet or not, there is an effect that the contents of the payload can be easily decoded from the packet header.

According to the present invention, when a packet including the digit number of the DTMF signal is lost, the DTMF signal is generated based on the result of decoding the digit number included in the packet received immediately before the lost packet. Normal DT even when a received packet is lost
There is an effect that an MF signal can be output.

According to the present invention, a predetermined number of consecutive audio data including the first output packet after the input signal becomes voiced, and the digit number of the DTMF signal and the audio data encoded by the audio encoding unit at a low bit rate. Is output, a packet containing only the digit number of the DTMF signal is output until a voiced end notification packet is output thereafter, so that the packet size can be reduced and transmission efficiency can be improved. There is an effect that can be.

According to the present invention, the receiving means calculates the signal level of the signal output by the audio decoding unit, and controls the receiving means to output a DTMF signal at a level equal to the calculated signal level. DTMF equal to the signal level of the DTMF signal input from the terminal device side
Since the signal can be output by the receiving means, the DTMF signal can be transmitted with higher accuracy.

According to the present invention, the receiving means calculates the mean square of the signal level of the signal output from the audio decoding unit, and sets the mean square value as the signal level. is there.

According to the present invention, the average value of the absolute value of the signal level of the signal output from the audio decoding unit is calculated by the receiving means, and the average value of the absolute value is used as the signal level. is there.

According to the present invention, the receiving unit calculates the maximum value of the signal level of the signal output from the audio decoding unit,
Since this maximum value is used as the signal level, the above paragraph 0087
Has the same effect as.

According to the present invention, the gain information included in the audio data obtained by encoding the received packet at a low bit rate is extracted, and the signal level of the signal output by the audio decoding unit is calculated based on the gain information. By controlling the signal level equal to the signal level calculated based on the gain information,
Since the TMF signal is output, the transmitting means has a DTMF substantially equal to the signal level of the DTMF signal input from the terminal device.
Since the receiving means can output the signal, there is an effect that the DTMF signal can be transmitted with higher accuracy.

According to the present invention, the transmitting means calculates the signal level of the input signal, and multiplexes the digit number of the DTMF signal with the calculated signal level, or the low bit rate coded audio data and DTM.
Since a packet obtained by multiplexing the digit number of the F signal and the calculated signal level is output, and the receiving means controls the signal level equal to the signal level extracted from the packet received from the transmitting means to output the DTMF signal.
As compared with the configuration of paragraph 086 to paragraph 0090, there is an effect that the receiving unit can more accurately reproduce the signal level of the DTMF signal input to the transmitting unit and generate the DTMF signal.

According to the present invention, the transmitting means calculates the mean square of the signal level of the input signal, and uses the mean square value as the signal level. Therefore, the same effect as in paragraph 0091 is obtained.

According to the present invention, the transmitting means calculates the average of the absolute value of the signal level of the input signal, and the average of the absolute values is used as the signal level.

According to the present invention, since the maximum value of the signal level of the input signal is calculated in the transmitting means and this maximum value is used as the signal level, the same effect as in paragraph 0091 is obtained.

According to the present invention, if the result of the determination as to whether or not the signal is a DTMF signal is not a DTMF signal until a predetermined time elapses after the input signal becomes sounded, then the sound is determined to be silent. Since the DTMF signal is not determined until the DTMF signal is detected, it is possible to prevent the DTMF detection unit from outputting a packet in which the low bit rate coded voice data and the DTMF number are multiplexed in the middle of a sound section due to an erroneous determination by the DTMF detection unit. There is an effect that an increase in the packet size can be prevented.

According to the present invention, when a packet containing only audio data that has been encoded at a low bit rate after the input signal is voiced, the low bit rate encoding is performed until a voiced end notification packet is output. Since only the packet of the converted audio data is output,
There is an effect that the influence of the erroneous determination of the TMF signal can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a voice packet communication device according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a format of a packet output by a packet assembling unit.

FIG. 3 is a diagram illustrating a relationship between a packet sequence received by a receiving unit of the voice packet communication device according to the first embodiment, an input of a packet decomposer, and a control signal input to an output selector.

FIG. 4 is a diagram showing a relationship between a packet sequence received by a receiving unit of the voice packet communication device according to the first embodiment, an input of a packet decomposer, and a control signal input to an output selector.

FIG. 5 is a diagram showing a relationship between a packet sequence received by a receiving unit of the voice packet communication device according to the first embodiment, an input of a packet decomposer, and a control signal input to an output selector.

FIG. 6 is a diagram illustrating a relationship between a packet sequence received by a receiving unit of the voice packet communication device according to the first embodiment, an input of a packet decomposer, and a control signal input to an output selector.

FIG. 7 is a diagram showing a relationship between a packet sequence received by a receiving unit of the voice packet communication device according to the first embodiment, an input of a packet decomposer, and a control signal input to an output selector.

FIG. 8 is a diagram showing a relationship between a packet sequence received by a receiving unit of the voice packet communication device according to the first embodiment, an input of a packet decomposer, and a control signal input to an output selector.

FIG. 9 shows a control signal input to a packet disassembling unit, a DTMF signal generating unit input, and a control signal input to an output selecting unit when a packet sequence received by the receiving unit of the voice packet communication device according to the first embodiment has a loss. FIG.

FIG. 10 is a diagram showing a packet format handled by a voice packet communication device according to a second embodiment of the present invention.

FIG. 11 shows a received packet sequence, an input of a packet decomposer, and a control signal input to an output selector when the receiving means of the voice packet communication device according to the second embodiment detects a DTMF signal continuously for 30 ms or more. FIG.

FIG. 12 is a diagram showing a configuration of a voice packet communication device according to a third embodiment of the present invention.

FIG. 13 is a diagram showing a configuration of a voice packet communication device according to a fourth embodiment of the present invention.

FIG. 14 is a diagram showing a configuration of a voice packet communication device according to a fifth embodiment of the present invention.

FIG. 15 is a diagram showing a packet format of the voice packet communication device according to the fifth embodiment.

FIG. 16 is a block diagram showing a configuration of a conventional voice / DTMF signal transmission device.

[Explanation of symbols]

1 DTMF detection section, 2 voice coding section, 3 voice / non-voice determination section, 4, 4A packet assembling section, 4a, 4c payload multiplexing section, 4b header addition section, 5 buffer section (buffer), 6 received signal type determination Unit, 7 packet readout control unit, 8, 8a packet decomposition unit, 9 audio decoding unit, 10, 10a, 10b, 10c DTMF signal generation unit, 11 silence generation unit, 12 output selection unit, 13 level calculation unit (received signal level calculation Unit), 14 gain information extraction unit, 15 level calculation unit (transmission signal level calculation unit).

Claims (20)

[Claims] A voice encoding unit that encodes an input signal at a low bit rate; a voice / non-speech determining unit that determines voice / non-voice of the input signal; and whether the input signal is a DTMF signal. And a DTMF detection unit that decodes the digit number if the input signal is a DTMF signal. The DTMF detection unit and the sound / non-speech determination unit determine that the input signal is a sound and a voice signal. And the audio encoding unit packetizes the audio data subjected to the low bit rate encoding and outputs the packetized data to a packet line. A sound end notification packet is output to the packet line, and the DTMF detection unit and the sound
If the input signal is determined to be voiced and a DTMF signal by the silence determination unit, the voice encoding unit outputs a packet in which the low bit rate coded voice data and the digit number of the DTMF signal are multiplexed to the packet line. Transmitting means having a packet assembling unit that performs a predetermined process on a packet in which the low bit rate encoded voice data and the digit number of the DTMF signal are multiplexed from the first packet in the sound period of the packet received from the packet line. If the received signal is received continuously, the voiced section is determined to be a DTMF signal; otherwise, the received signal type determination unit determines to be an audio signal. A buffer for storing the received packet, and the voiced section is determined to be an audio signal by the received signal type determination unit. A voice decoding section for decoding the audio data low bit rate coding included in the signal packet, DTMF the voiced section is included in the received packet is determined to be a DTMF signal by the receiving signal type determination unit
And a receiving unit having a DTMF signal generating unit for generating a DTMF signal based on the digit number of the signal. 2. The voice packet communication device according to claim 1, wherein the DTMF detection unit detects whether or not the input signal is a DTMF signal in a time equal to or less than a coding delay of the voice coding unit. . 3. The reception signal type determination unit determines that the signal type is a voice signal when the determination result of the signal type of the voiced section is not output at the time of receiving the voice termination notification packet. The voice packet communication device according to claim 1. 4. The received signal type determination unit does not output a signal type determination result of a voiced section even if a predetermined time or more has passed since the first packet was received from the packet line after being voiced. 2. The voice packet communication device according to claim 1, wherein the signal type is determined as a voice signal. 5. A packet assembler, comprising: a packet containing only audio data encoded at a low bit rate;
Depending on whether the packet contains only the decoding result of the digit number of the MF signal or the packet in which the low bit rate encoded voice data and the decoding result of the digit number of the DTMF signal are multiplexed, different packet headers are used. 2. The voice packet communication device according to claim 1, wherein the voice packet communication device is added. 6. A DTMF signal generator, when a packet including a digit number of a DTMF signal is lost, generates a DTMF signal based on a result of decoding a digit number included in a packet received immediately before the lost packet. The voice packet communication device according to claim 1, wherein 7. A packet assembler, comprising: a predetermined number of packets including a packet to be output first after an input signal becomes a sound; 2. The voice packet communication apparatus according to claim 1, wherein when a packet multiplexed with a number is output, a packet including only the digit number of the DTMF signal is output until a voiced end notification packet is output thereafter. . 8. The reception means includes a reception signal level calculation section for calculating a signal level of a signal output from the audio decoding section, and the DTMF signal generation section has a level equal to the signal level calculated by the reception signal level calculation section. 2. The voice packet communication apparatus according to claim 1, wherein the apparatus outputs a DTMF signal under control. 9. A reception signal level calculation unit calculates a mean square of signal levels of signals output from the audio decoding unit, and calculates
9. The method according to claim 8, wherein the square mean value is used as the signal level.
The voice packet communication device according to claim 1. 10. The reception signal level calculator according to claim 8, wherein the average of the absolute values of the signal levels of the signals output from the audio decoder is calculated, and the average of the absolute values is used as the signal level. Voice packet communication device. 11. The voice packet communication according to claim 8, wherein the reception signal level calculation unit calculates a maximum value of a signal level of the signal output from the voice decoding unit, and sets the maximum value as a signal level. apparatus. 12. A receiving means for extracting gain information included in low bit rate encoded audio data of a received packet and calculating a signal level of a signal output by an audio decoding unit based on the gain information. 2. The voice packet communication device according to claim 1, further comprising an information extraction unit, wherein the DTMF signal generation unit outputs the DTMF signal by controlling the signal level to be equal to the signal level calculated by the gain information extraction unit. 13. The transmission means includes a transmission signal level calculator for calculating a signal level of an input signal, and the packet assembler multiplexes the digit number of the DTMF signal and the signal level calculated by the transmission signal level calculator. Or a packet obtained by multiplexing the low bit rate encoded voice data, the digit number of the DTMF signal and the signal level calculated by the transmission signal level calculation unit, and outputs the DTMF signal from the transmission unit 2. The voice packet communication device according to claim 1, wherein the DTMF signal is output by controlling to a level equal to a signal level extracted from a received packet. 14. The voice packet communication device according to claim 13, wherein the transmission signal level calculation unit calculates a root-mean-square of the signal level of the input signal, and uses the root-mean-square value as the signal level. 15. The voice packet communication device according to claim 13, wherein the transmission signal level calculator calculates an average of the absolute value of the signal level of the input signal, and sets the average of the absolute value as the signal level. 16. The voice packet communication device according to claim 13, wherein the transmission signal level calculation unit calculates the maximum value of the signal level of the input signal, and sets the maximum value as the signal level. 17. The DTMF detection unit, if the determination result as to whether or not the signal is a DTMF signal determines that the signal is not a DTMF signal until a predetermined time has elapsed after the input signal has sound, and thereafter, the DTMF detection unit determines that there is no sound. 2. The voice packet communication device according to claim 1, wherein the DTMF signal is not determined until the condition is satisfied. 18. A packet assembler, when outputting a packet including only low bit-rate encoded voice data after an input signal becomes voiced, outputs a low bit-rate encoded packet until a voiced end notification packet is output thereafter. 2. The method according to claim 1, wherein only packets of the converted audio data are output.
The voice packet communication device according to claim 1. 19. A voice packet communication method for generating a packet based on an input signal from a terminal device, outputting the packet to a packet line, and receiving the packet from the packet line, wherein the voice signal for encoding the input signal at a low bit rate is provided. Encoding step, if the input signal is a voice and a voice signal, the voice data encoded at a low bit rate in the voice coding step is packetized and output to a packet line; When the input signal changes to silence, a sound end notification packet is output to the packet line.
A packet assembling step of outputting, to the packet line, a packet obtained by multiplexing voice data encoded at a low bit rate in the voice encoding step and a digit number of the DTMF signal as a TMF signal; When a packet in which the low bit rate encoded voice data and the digit number of the DTMF signal are multiplexed from the head packet of the voice section of the packet received from the packet line for a predetermined number of times, the voice section is recognized. A received signal type determining step of determining a DTMF signal and otherwise determining an audio signal; and a low signal level included in a received packet whose sound section is determined to be an audio signal in the received signal type determining step. An audio decoding step of decoding bit-rate encoded audio data; Sound interval is DTMF in step
DTM included in the received packet determined to be a signal
A receiving step having a DTMF signal generating step of generating a DTMF signal based on the digit number of the F signal. 20. The voice packet communication method according to claim 19, wherein it is detected whether or not the input signal is a DTMF signal in a time equal to or shorter than the coding time in the voice coding step.