JP2004158963A - Tone quality evaluation system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a tone quality evaluation system capable of evaluating tones in test conditions close to an actual environment and having reproducibility. <P>SOLUTION: The tone quality evaluation system for voice communication via a network includes at least two sound proof chambers for housing each of voice communication terminals which perform voice communication with each other via the network. Each of the sound proof chambers includes an input signal generating device for transmitting an input signal to each of the voice communication terminals, and an evaluation device for receiving an output signal from each of the voice communication terminals to output a result of evaluation for the output signal. The transmission of the input signal, the reception of the output signal and/or the evaluation of the output signal are individually controlled by a control device. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a sound quality evaluation device that performs sound quality evaluation on a voice communication system via a network such as VoIP (Voice over Internet Protocol) on the Internet.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, sound quality evaluation of a sound quality evaluation device for a voice communication system via a network is performed by directly connecting a telephone to the device and listening to an actual call of a speaker, or connecting a dedicated measuring device to the measuring device. This is implemented by inputting a reference voice from a computer and evaluating the output.
[0003]
[Problems to be solved by the invention]
However, in the conventional method in which a human speaker talks, it is difficult to accurately reproduce the volume and timing of the utterance, and there is a problem that the reproducibility of the evaluation is poor. In addition, evaluation is often performed in the same room, and the voice of the speaker leaks directly to the telephone receiver of the other party, so that the evaluation can only be performed in a state that cannot be said to be an actual call state. There are points.
[0004]
On the other hand, in the method using the reference sound, only the reference sound is directly input from the measuring instrument, so that background noise that would affect the sound quality in the actual use environment is not input, and realistic sound quality evaluation is performed. In addition, there is a problem that it is not possible to reproduce a realistic change in network load in voice communication using packet communication such as VoIP in a test environment.
[0005]
The present invention has been made in view of the above-described problems, and an object of the present invention is to provide a sound quality evaluation device that can perform sound quality evaluation under test conditions that are close to an actual use environment and have reproducibility.
[0006]
[Means for Solving the Problems]
A sound quality evaluation device according to the present invention is a sound quality evaluation device for voice communication via a network, and includes at least two soundproof rooms accommodating at least two voice communication terminals that perform voice communication with each other via the network. An input signal generator accommodated in each of the soundproof rooms and transmitting an input signal to each of the voice communication terminals; and an output signal received and output from each of the sound communication terminals accommodated in each of the soundproof rooms. It is characterized by including an evaluation device that outputs an evaluation result for a signal, and a control device that individually controls transmission of the input signal, reception of the output signal, and / or evaluation of the output signal.
[0007]
【Example】
Embodiments of the present invention will be described in detail with reference to the accompanying drawings.
FIG. 1 shows an embodiment of the present invention, and shows an overall view of a sound quality evaluation device 100. The sound quality evaluation device 100 includes two soundproof rooms, that is, an A-side soundproof room 103 and a B-side soundproof room 104, and a control room 105. The A-side soundproof room 103 and the B-side soundproof room 104 are connected by an IP network 106 to enable voice communication between the evaluation target devices, which are voice communication terminals housed inside each. The A-side soundproof room 103 and the B-side soundproof room 104 are rooms that block external sound. One soundproof room is divided into two soundproof rooms by soundproof walls, that is, the A-side soundproof room 103 and the B-side soundproof room 104. May be realized by dividing into
[0008]
The control room 105 controls a device housed in each soundproof room by a control signal by a device housed therein. Then, the control room 105 also receives the signal of the evaluation result from the A-side soundproof room 103 and the B-side soundproof room 104.
FIG. 2 shows a detailed configuration of the sound quality evaluation device 100 shown in FIG.
The A-side soundproof room 103 includes an A-side input signal generation device 111, an A-side evaluation target device 112, an A-side evaluation device 113, and an A-side background noise generation device 114. The B-side soundproof room 104 includes a B-side input signal generation device 121, a B-side evaluation target device 122, a B-side evaluation device 123, and a B-side background noise generation device 124. The control room 105 includes a control device 131, a load generating device 132, and an evaluation result recording device 133. The IP network 106 is connected between the A-side evaluation target device 112 and the B-side evaluation target device 122.
[0009]
Referring to the A-side soundproof room 103, the A-side input signal generation device 111 has a function of generating an evaluation input signal to the A-side evaluation target device 112 and emitting sound, and outputs an input signal including voice. A sound source for generating the sound is provided, and reproduction, stop, and volume control can be controlled by an external control signal. As an example of the A-side input signal generator 111, devices such as a tape recorder, a DAT player, a CD player, an MD player, and a personal computer (PC) can be used. As an input signal, that is, an input signal for evaluation, a known reference audio signal provided by an international standardization organization such as ITU-T, an audio signal in which an actual call is recorded, or a pseudo-created audio data signal, or Use other appropriate criteria signals.
[0010]
The A-side evaluation target device 112 is a voice communication terminal to be subjected to sound quality evaluation, and is, for example, a VoIP device such as an IP telephone. The input signal from the A-side input signal generation device 111 is emitted to the soundproof room 103 and then propagates through the room, and then is taken into the A-side evaluation target device 112 as a sound input. Note that the input signal may be taken in directly as an audio input via a cable.
[0011]
The A-side evaluation device 113 has a function of taking in an output signal of the A-side evaluation target device 112, that is, an audio output, and performing sound quality evaluation. The audio output may be captured from the A-side evaluation target device 112 via a cable, or may be propagated through the soundproof room and then captured by a microphone or the like. As a sound quality evaluation method, ITU (International Telecommunication Union) -T Recommendation P. The sound quality may be evaluated using PSQM (Perceptual Speech Quality Measurement) 861, but the A-side evaluation device 113 is only required to be a device capable of evaluating sound quality, and the evaluation function based on such a standard is not limited. By the way, PSQM is a method of objectively assessing the quality of speech degraded by telephony networks, relating the range of distortion types to subjective quality, and focusing on different coding types and transmission errors. This is a method of network test. ITU-T Recommendation P. The definition of 861 mainly defines speech compression, digital speech insertion, and packetization. The result index obtained by PSQM gives a value from 0 to 6.5, with 0 indicating very good speech quality and 6.5 indicating very poor speech quality.
[0012]
The A-side background noise generator 114 has a function of generating background noise around the A-side evaluation target device 112, includes a sound source that generates the background noise, and can control reproduction, stop, and volume control from outside. And For example, a device such as a tape recorder, a DAT player, a CD player, an MD player, or a personal computer (PC) can be used. As background noise, noise such as office noise and environmental noise can be used. Even if the background noise is emitted as sound input to the A-side evaluation target device 112 after being emitted into the soundproof room 103 as shown in the figure, the A-side evaluation target device 112 is directly input as a sound input via a cable. May be taken in.
[0013]
Referring to the B-side soundproof room 104, the B-side input signal generation device 121 has a function of generating an input signal for evaluation to the B-side evaluation target device 122 and emitting sound, and outputs an input signal including voice. A sound source for generating the sound is provided, and reproduction, stop, and volume control can be controlled by an external control signal. Examples of the B-side input signal generator 121 include devices such as a tape recorder, a DAT player, a CD player, an MD player, and a personal computer (PC). As an input signal, that is, an input signal for evaluation, a known reference audio signal provided by an international standardization organization such as ITU-T, an audio signal in which an actual call is recorded, or a pseudo-created audio data signal, or Use other appropriate criteria signals.
[0014]
The B-side evaluation target device 122 is a voice communication terminal to be subjected to sound quality evaluation, and is, for example, a VoIP device such as an IP telephone. The input signal from the B-side input signal generator 121 is emitted to the soundproof room 104 and then propagates through the room, and then is taken into the B-side evaluation target device 122 as a voice input. Note that the input signal may be taken in directly as an audio input via a cable.
[0015]
The B-side evaluation device 123 has a function of taking in an output signal of the B-side evaluation target device 122, that is, an audio output, and performing sound quality evaluation. The audio output may be captured from the B-side evaluation target device 122 via a cable, or may be propagated through the soundproof room and then captured by means such as a microphone. As a sound quality evaluation method, a method similar to that in the A-side evaluation device 113 is usually used.
[0016]
The B-side background noise generation device 124 has a function of generating background noise around the B-side evaluation target device 122, includes a sound source that generates the background noise, and can control reproduction, stop, and volume adjustment from outside. And For example, a device such as a tape recorder, a DAT player, a CD player, an MD player, or a personal computer (PC) can be used. As background noise, noise such as office noise and environmental noise can be used. Even if the background noise is emitted into the soundproof room 104 as shown in the drawing, and is taken into the B-side evaluation target device 122, the background noise is taken into the B-side evaluation target device 122 as a voice input directly via a cable. Good.
[0017]
Referring to the control room 105, the control device 131 includes devices used for evaluation, that is, an A-side input signal generation device 111, an A-side evaluation target device 112, an A-side evaluation device 113, and an A-side background noise generation device. Control of signal generation or signal stop for each device including 114, B-side input signal generation device 121, B-side evaluation target device 122, B-side evaluation device 123, and B-side background noise generation device 124 Generate signals to control them. The control device 131 also transmits a control signal including load information to the load generating device 132 to fluctuate the network load, and controls the operation of the evaluation result recording device 133 by the control signal. The load generating device 132 is a device that artificially applies an arbitrary load to the IP network 106 according to a control signal from the control device 131. The evaluation result recording device 133 is a device that records results evaluated by the A-side evaluation device 113 and the B-side evaluation device 123 in response to a control signal from the control device 131 for later result analysis.
[0018]
In the present embodiment, since the VoIP device is assumed as the device to be evaluated, the A-side evaluation device 112 and the B-side evaluation device 122 are connected via the IP network 106. In the case of a normal telephone set other than the above, a normal line such as a telephone line can be selected for connection with the A-side evaluation target device 112 and the B-side evaluation target device 122.
[0019]
With reference to FIG. 2, a control flow in the sound quality evaluation device 100 according to the first embodiment will be described. The order of this control is an example, and the present invention is not limited to such order.
The control device 131 transmits the evaluation result recording start signal start_save to the evaluation result recording device 133 to start the recording operation. Next, the control device 131 transmits an evaluation start signal start_meas to the A-side evaluation device 113. The A-side evaluation device 113 that has received the evaluation start signal start_meas starts the sound quality evaluation operation, and sequentially reports the A-side evaluation result measure_A to the evaluation result recording device 133. The A-side evaluation result measure_A uses the above-described PSQM value.
[0020]
Next, the control device 131 transmits an evaluation start signal start_meas to the B-side evaluation device 123. The B-side evaluation device 123 that has received the evaluation start signal starts the sound quality evaluation operation, and sequentially reports the B-side evaluation result measure_B to the evaluation result recording device 133. The B-side evaluation result measure_B uses the PSQM value in the same manner as the A-side evaluation result measure_A. Note that the A-side evaluation result measure_A and the B-side evaluation result measure_B may be the sound quality evaluation values themselves, and additional information may be added as needed.
[0021]
The evaluation result recording device 133 records the A-side evaluation result measure_A and the B-side evaluation result measure_B on a storage medium such as a memory or a disk.
Next, the control device 131 transmits the background noise generation signal start_noise to the A-side background noise generation device 114 and the B-side background noise generation device 124, respectively. The A-side background noise generator 114 and the B-side background noise generator 124 that have received the background noise generation signal generate the background noise as A-side background noise noise_A and B-side background noise noise_B, respectively. Then, the A-side background noise noise_A is emitted into the B-side soundproof room 104 and the B-side background noise noise_B is emitted. Note that it is also possible to perform sound quality evaluation in a noise-free state in which background noise is not transmitted.
[0022]
Next, the control device 131 transmits a call connection signal ctrl_connect to the A-side evaluation target device 112 and the B-side evaluation target device 122. As a result, the A-side evaluation target device 112 and the B-side evaluation target device 122 are connected via the IP network 106.
Next, the control device 131 transmits the input signal generation signal start_signal to the A-side input signal generation device 111. Upon receiving the input signal generation signal start_signal, the A-side input signal generator 111 generates an input signal for evaluation as an A-side input signal signal_A, and inputs the A-side background noise noise_A to the A-side evaluation target device 112 as a voice input. I do. Next, the control device 131 transmits the input signal generation signal start_signal to the B-side input signal generation device 121. Upon receiving the input signal generation signal start_signal, the B-side input signal generation device 121 generates an input signal for evaluation as a B-side input signal signal_B, and inputs the signal to the B-side evaluation target device 122 as a voice input together with the B-side background noise noise_B. . As a result, a conversation during a call is simulated between the A-side evaluation target device 112 and the B-side evaluation target device 122.
[0023]
The evaluation can be ended at an arbitrary timing by performing the following procedure to end the evaluation. First, the control device 131 transmits an input signal stop signal stop_signal to the A-side input signal generation device 111. In response to the input signal stop signal stop_signal, the A-side input signal generator 111 stops generating the A-side input signal signal_A. Next, by transmitting a call disconnection signal ctrl_disconnect to the A-side evaluation target device 112, the extension call of the A-side evaluation target device 112 is disconnected, and further, a background noise stop signal stop_noise is transmitted to the A-side background noise generation device 114. Accordingly, the A-side background noise generator 114 stops generating the A-side background noise noise_A. Further, the control device 131 transmits the evaluation stop signal stop_meas to the A-side evaluation device 113, so that the A-side evaluation device 113 stops the sound quality evaluation operation. The operation of each device of the B-side soundproof quality is stopped by the same procedure. The control device 131 also transmits an evaluation result recording stop signal stop_save to the evaluation result recording device 133, and the evaluation result recording device 133 stops the operation of recording the evaluation result.
[0024]
During the evaluation operation described above, the control device 131 may control the load on the IP network 106 by transmitting load information ctrl_traf including designation information such as a load factor to the load generating device 132. In this case, the load generator 132 increases the network load of the IP network 106 by transmitting a dummy packet based on the load information ctrl_traf. As described above, in the first embodiment, by using the soundproof room, the crosstalk of the sound between the two terminals that make communication is prevented, and the same state as the actual call state can be reproduced. In addition, since the load on the IP network can be arbitrarily set, it is possible to completely distinguish and evaluate the effect on the sound quality given by the performance of the terminal and the effect on the sound quality due to packet delay or packet loss.
[0025]
Further, since the voice of the speaker and the background noise are input as completely different settings, the state of the telephone user in various usage forms can be easily reproduced. Since the input signal for evaluation and the generation of the background noise are centrally managed by the control device, the generation timing can be arbitrarily set or reproduced in an ordered manner. The reproducibility required for sound quality evaluation is guaranteed. Further, by using the A-side input signal generator and / or the B-side input signal generator as a device such as a tape recorder capable of reproducing real audio, evaluation using actual audio is also possible.
<Second embodiment>
FIG. 3 shows a second embodiment, and shows a detailed configuration of the sound quality evaluation device 100.
[0026]
In the second embodiment, a function of generating a control signal for adjusting the volume of the background noises noise_A and noise_B and the input signals signal_A and signal_B is added to the control device 131 having the configuration shown in the first embodiment. The other configuration is the same as that in the first embodiment, and thus the description is omitted.
The operation of the sound quality evaluation device 100 according to the second embodiment will be described with reference to FIG. The order of this control is an example, and the present invention is not limited to such order.
[0027]
The control device 131 transmits the background noise volume adjustment signal ctrl_noise including the volume designation information to the A-side background noise generation device 114. The A-side background noise generator 114 adjusts the volume of the A-side background noise noise_A based on the background noise volume control signal ctrl_noise. Next or simultaneously, the control device 131 transmits the background noise volume adjustment signal ctrl_noise including the volume designation information to the B-side background noise generation device 124. The B-side background noise generator 124 adjusts the volume of the B-side background noise noise_B based on the background noise volume adjustment signal ctrl_noise.
[0028]
Further, the control device 131 transmits an input signal volume control signal ctrl_signal including the volume specifying information to the A-side input signal generating device 111. The A-side input signal generator 111 adjusts the volume of the A-side input signal signal_A based on the input signal volume control signal ctrl_signal. Next or simultaneously, the control device 131 transmits an input signal volume control signal ctrl_signal including the volume designation information to the B-side input signal generation device 121. The B-side input signal generator 121 adjusts the volume of the B-side input signal signal_B based on the input signal volume control signal ctrl_signal.
[0029]
FIG. 4 shows a time chart of a control example executed by the control device 113 shown in FIG. As shown in this time chart, the control device 113 in FIG. 3 controls each device. For example, the evaluation result recording device 133 starts operation in response to the recording result recording start signal start_save, and ends operation in response to the recording result recording stop signal stop_save. The evaluation devices 113 and 123 start operation in response to the evaluation start signal start_meas, and end operation in response to the evaluation stop signal stop_meas. In response to the background noise generation signal start_noise, the background noise generation devices 114 and 124 start generating background noise, change the volume in accordance with the background noise volume adjustment signal ctrl_noise, and change the volume in response to the background noise stop signal stop_noise. Stop the occurrence. The evaluation target devices 112 and 122 establish a call connection to the IP network 106 according to the call connection signal ctrl_connect, and disconnect the call according to the call disconnection signal ctrl_disconnect. The load generator 132 changes the network load according to the network load adjustment signal ctrl_traf. In response to the input signal generation signal start_signal, the input signal generation devices 111 and 121 start generating the input signal, adjust the volume according to the input signal volume adjustment signal ctrl_signal, and input the signal according to the input signal stop signal stop_signal. Stop the occurrence of The order of control by the control device 131 in the figure is an example, and various sound quality evaluation conditions can be set in various orders.
[0030]
As described above, in the second embodiment, since the background noise and the input signal can be controlled at an arbitrary volume at an arbitrary timing, the evaluation environment, for example, evaluation items such as a noise-to-signal ratio and a sound level are actually used. The environment can be set in accordance with the state, and the evaluation in the actual use state, which was very difficult to reproduce and measure in the past, can be easily performed.
<Third embodiment>
FIG. 5 shows a detailed configuration of the sound quality evaluation device 100 according to the third embodiment of the present invention. The A-side soundproof room 103 includes an A-side input signal generation device 111, an A-side evaluation target device 112, an A-side evaluation device 113, and an A-side background noise generation device 114 as in the first embodiment. . In the third embodiment, the A-side evaluation target device 112 is connected to the IP network 106 to enable telephone communication by VoIP, and the PBX 119 connected to the VoIP gateway 117 to exchange telephone communication signals. And a plurality of key telephones 118a to 118c each connected to the PBX 119 and enabling extension calls between channels. In each of the plurality of key telephones 118a to 118c, the input signal signal_A for evaluation and the background noise noise_A from the input signal generating device 111 are input as voice inputs, and the voice output measure_A from them is taken into the evaluation device 113. Although three key telephones 118a to 118c are shown in FIG. 5, the number is an example, and the number of telephones can be many.
[0031]
The operation of the sound quality evaluation device 100 according to the third embodiment will be described with reference to FIG. The order of this control is an example, and the present invention is not limited to such order.
The control device 131 starts the operation of each of the evaluation result recording device 133 and the A-side evaluation device 113 using the control signal. The control device 131 may operate the A-side background noise generation device 114 using the control signal to generate the background noise noise_A, or may be in a noiseless state in which the background noise noise_A is not transmitted.
[0032]
Next, the control device 131 transmits a call connection signal ctrl_connect to the A-side evaluation target device 112. In accordance with this call connection signal ctrl_connect, an extension connection is established between any two of the key telephones 118a to 118c of the A-side evaluation target device 112, for example, between the key telephones 118a and 118b. Then, the control device 131 transmits the input signal generation signal start_signal to the A-side input signal generation device 111. In response to this input signal generation signal start_signal, the A-side input signal generation device 111 generates an evaluation signal as an A-side input signal signal_A, and inputs the signal to the key telephones 118a and 118b of the A-side evaluation target device 112, respectively. Different signals for evaluation may be used for each channel used for the extension call, or the same signal may be used. By the above operation, the pseudo conversation during the extension call is recorded in the A-side evaluation target device 112.
[0033]
As described above, in the third embodiment, the influence of the IP network can be ignored since the telephone communication path is turned back by the VoIP gateway 117, and the sound quality of the voice communication terminal itself can be evaluated under the condition with reproducibility.
In the first and second embodiments, VoIP devices are evaluated. However, if the IP network 106 in these embodiments is changed to a normal telephone line, the sound quality evaluation device according to the present invention can be used for a fixed telephone, The present invention is also applicable to sound quality evaluation of other voice communication systems such as mobile phones.
In the first embodiment and the second embodiment, the method for evaluating the communication between two terminals by using two soundproof rooms has been described. However, three or more sets of the soundproof room and each device in the soundproof room are prepared. Then, the present invention is also applicable to sound quality evaluation for simultaneous communication between a plurality of terminals of three or more terminals. It should be noted that the soundproof room in the present disclosure does not need to allow humans to enter and exit, and may be a space that is soundproof and shielded and in which necessary devices can be installed.
[0034]
【The invention's effect】
As described above, the sound quality evaluation device according to the present invention can add background noise in a soundproof test environment, and further provides a test environment closer to an actual use environment by giving a network load. In addition, since the evaluation target and each device for evaluation are collectively controlled by the control device, the sound quality can be evaluated under reproducible test conditions.
[Brief description of the drawings]
FIG. 1 is a perspective view illustrating an appearance of a sound quality evaluation device according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a detailed configuration of a voice evaluation device according to a first embodiment of the present invention.
FIG. 3 is a block diagram illustrating a detailed configuration of a voice evaluation device according to a second embodiment of the present invention.
FIG. 4 is a time chart of a control example executed by the control device 113 shown in FIG. 3;
FIG. 5 is a block diagram showing a detailed configuration of a voice evaluation device according to a third embodiment of the present invention.
[Explanation of symbols]
Reference Signs List 100 sound evaluation device 103 A-side soundproof room 104 B-side soundproof room 105 control room 106 IP network 111, 121 input signal generation device 112, 122 evaluation target device 113, 123 evaluation device 114, 124 background noise generation device 115, 116 volume control Part 117 VoIP gateway 118a to 118c Button telephone 119 PBX
131 control device 132 load generation device 133 evaluation result storage device

Claims (6)

A sound quality evaluation device for voice communication via a network,
At least two soundproof rooms each accommodating at least two voice communication terminals that make voice communication with each other via the network;
An input signal generation device accommodated in each of the soundproof rooms and transmitting an input signal to each of the voice communication terminals,
An evaluation device housed in each of the soundproof rooms, receiving an output signal from each of the voice communication terminals and outputting an evaluation result for the output signal,
A control device for individually controlling transmission of the input signal, reception of the output signal and / or evaluation of the output signal;
A sound quality evaluation device comprising: 2. The sound quality evaluation device according to claim 1, wherein the input signal is a reference voice signal, a voice signal obtained by recording an actual call, or pseudo-created voice data. 3. The sound quality evaluation device according to claim 1, wherein the evaluation result includes an evaluation index based on PSQM (Perceptual Speech Quality Measurement). 4. The apparatus further includes at least two background noise generators housed in each of the soundproof rooms and transmitting a background noise signal to the voice communication terminal, wherein the control device controls transmission of the background noise signal. The sound quality evaluation device according to claim 1. The sound quality evaluation device according to any one of claims 1 to 4, wherein the control device further controls the volume of the input signal and / or the volume of the background noise signal individually. 6. The apparatus according to claim 1, further comprising means for causing a change in transmission delay time of a packet for voice communication and / or a packet loss in the network to change a load on the network. The described sound quality evaluation device.

Images