JP2005184220A - Voice communication apparatus and voice quality estimation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To inexpensively estimate voice quality without the need for a large-scale system configuration in the case of making a speech by transmitting / receiving a voice packet to / from a speech opposite apparatus being a speech opposite party via a network. <P>SOLUTION: A voice communication apparatus for making a speech by transmitting / receiving a voice packet to / from the speech opposite apparatus being the speech opposite party via the network includes: a detection means for detecting a state of the network located between the voice communication apparatus and the speech opposite apparatus; an estimation means for estimating the voice quality on the basis of the network state detected by the detection means; and an output means for outputting information denoting the voice quality estimated by the estimation means; and also a recording means for recording a transmission timing every time an echo request packet is transmitted and recording a reception timing every time an echo response packet is received, and the detection means detects the state of the network on the basis of the recording contents. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a technique for estimating sound quality in a voice communication apparatus for performing a call by transmitting and receiving voice packets to and from a call partner apparatus that is a call partner side via a network.

Conventionally, a network evaluation system for evaluating network performance based on an arbitrary evaluation criterion has been proposed (see, for example, Patent Document 1).
JP 2002-164891 A

  However, this conventional network evaluation system evaluates the required value input from the input device and the actual measurement value obtained as a result of the traffic measurement in the measurement device based on an arbitrary evaluation criterion. There is a problem that the system configuration becomes relatively large and the sound quality cannot be evaluated at a low cost.

  An object of the present invention is to reduce the sound quality without requiring a large-scale system configuration when performing a call by sending and receiving voice packets to and from a call partner apparatus that is a call partner side via a network. It is to provide a technique for estimation.

  The present invention has been made to solve the above-described problems, and is a voice communication device for performing a call by transmitting and receiving voice packets to and from a call partner device that is a call partner side via a network. Detecting means for detecting the state of the network with the other party device, estimating means for estimating the sound quality based on the state of the network detected by the detecting means, and the sound quality estimated by the estimating means Output means for outputting the information to be expressed.

  According to the present invention, when a call is made by sending and receiving a voice packet to and from a call partner apparatus on the call partner side via a network, the network state is detected, and the detected network state is set. Since the sound quality is estimated based on the sound quality, the sound quality can be estimated at a low cost without requiring a large-scale system configuration.

  In the voice communication device, for example, an echo request packet is transmitted (for example, to the call partner device), and an echo response packet returned from the partner device (for example, the call partner device) that has received the echo request packet is transmitted. Receiving / transmitting means; and recording means for recording the transmission timing each time the echo request packet is transmitted and recording the reception timing each time the echo response packet is received; and the detection means The network status is detected based on the content recorded by the recording means.

  This shows an example of detecting the state of the network, and the detection means of the present invention is not limited to this. As described above, the network status is detected using the echo request packet and the echo response packet, and the sound quality is estimated based on the detected network status, so that a large-scale system configuration is not required and the cost is low. The sound quality can be estimated from one side.

  In the voice communication device, for example, the state of the network detected by the detection unit is such that a packet transmitted from the own device arrives at a partner device (for example, the partner device) via the network, It is represented by a value corresponding to a round trip time until a packet returned from the partner apparatus (for example, the communication partner apparatus) in response to the packet arrives at the own apparatus.

  This is an example of the state of the network detected by the detection means. The network state detected in the present invention is not limited to this. For example, the network state detected by the detection unit may be represented by a value corresponding to a packet loss rate of a packet transmitted / received by the transmission / reception unit.

The present invention can also be specified as a method invention as follows.
A method for estimating sound quality in a voice communication apparatus for performing a call by transmitting and receiving voice packets to and from a call partner apparatus that is a call partner side, A detection step for detecting a network state of the network, an estimation step for estimating sound quality based on the network state detected by the detection step, and an output step for outputting information representing the sound quality estimated by the estimation step; A sound quality estimation method comprising:

  In the sound quality estimation method, for example, an echo request packet is transmitted (for example, to the call partner device), and an echo response packet returned from the partner device (for example, the call partner device) that has received the echo request packet is transmitted. A transmission / reception step for receiving, and a recording step for recording the transmission timing each time the echo request packet is transmitted, and a recording step for recording the reception timing each time the echo response packet is received, the detection step further comprising: The network status is detected based on the content recorded by the recording means.

  In the sound quality estimation method, for example, the network state detected by the detection step is such that a packet transmitted from the own device arrives at the partner device (for example, the other party device) via the network, It is represented by a value corresponding to a round trip time until a packet returned from the partner apparatus (for example, the communication partner apparatus) in response to the packet arrives at the own apparatus.

  The network state detected by the detecting step is represented by a value corresponding to the packet loss rate of the packet transmitted / received by the transmitting / receiving step.

  According to the present invention, when a call is made by transmitting and receiving voice packets to and from a call partner device that is a call partner side, a large-scale system configuration is not required and sound quality is reduced at a low cost. It is possible to estimate.

Hereinafter, a communication system including a voice communication apparatus according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram of a voice communication apparatus according to an embodiment of the present invention.
(Configuration of voice communication device)
The voice communication apparatus 100 is a communication apparatus having a function of realizing a call by transmitting and receiving voice packets to and from a call partner apparatus that is a call partner side by VoIP (Voice over IP). For example, when configuring a phone-to-phone type IP telephone in which an existing general telephone is connected to an IP network via a modem with a VoIP function, the voice communication apparatus 100 is connected to the modem with the VoIP function. It corresponds to. Further, when configuring a PC to phone type IP phone connected to an IP network via an information processing device such as a general personal computer, the information processing device such as the personal computer corresponds to the voice communication device 100. To do. Further, when configuring an IP telephone with a built-in VoIP function, the telephone itself corresponds to the voice communication apparatus 100.

  As shown in FIG. 1, the voice communication apparatus 100 includes an input function 101, an echo packet transmission function 102, an echo packet reception function 103, a transmission / reception status recording function 104, a transmission / reception record evaluation function 105, an evaluation result output function 106, And an audio codec (codec) information storage function 107 and the like. These functions are realized by predetermined software or a circuit.

The voice communication apparatus 100 also has a communication function (not shown), and is connected to the IP network via this communication function. The voice communication apparatus 100 also includes a codec (usually a plurality) (not shown). The codec realizes a decoding function and an encoding function. Note that the codec is an encoding / decoding device or device for digitizing audio that is an analog signal. The audio codec information storage function 107 is for storing the correspondence between codecs and their code rates such as bit rate and transmission interval. Other functions will be clarified by the following operation description.
(Operation of voice communication device)
Next, the operation of the voice communication apparatus 100 configured as described above will be described with reference to the drawings. FIG. 2 is a flowchart for explaining the operation of the voice communication apparatus 100.
(Sound quality estimation process overview)
In this embodiment, a so-called ping echo request packet is used to determine the state of the network (for example, represented by a delay) with the other device, and a known sound quality estimation method for this delay and other values. The sound quality is estimated by applying (EIF method).

  As shown in FIG. 2, it is assumed that a communication partner apparatus (for example, a voice communication apparatus 100 on the other party scheduled to make a call) is designated (S100). This is designated by the user via the input function 101, for example. When the other party device is designated, the voice communication device 100 starts to transmit an echo request packet (also referred to as an echo packet) to the designated other party device under preset conditions (S101). This is performed by the echo packet transmission function 102, for example.

  Since ping / ICMP is used, S100 may be a simple PC (information processing apparatus such as a personal computer) / router. Therefore, simple evaluation before mounting is also possible.

  The preset conditions are packet length, transmission interval, and the like. This condition is set by the user via the input function 101, for example. Alternatively, it can be set as follows. First, the correspondence between the codec and the codec information (packet length, transmission interval, etc.) is held in the storage device (voice codec information storage function 107) of the voice communication apparatus 100 (for example, in the form of a table). Then, when the codec to be used in the voice communication apparatus 100 is determined, the codec information corresponding to the codec scheduled to be used is acquired (read out) from the corresponding relationship and set.

Alternatively, the user is allowed to input the packet length via the input function 101, specify the bit rate, calculate the transmission interval based on these, and set them.
As described above, conditions such as packet length, transmission interval, and bit rate can be set in advance.

  The voice communication device 100 continuously transmits the echo request packet to the communication partner device specified in S100 according to preset conditions, that is, the set packet length, transmission interval, and other conditions (S101). . The number of the packet is stored in the contents of the echo request packet to be transmitted. FIG. 3 shows a format of an echo request packet (the same applies to an echo response packet described later). This is an ICMP (Internet Control Message Protocol) message format. The sequence number is set to a value that increases by 1 each time an echo request packet is transmitted. That is, the sequence number indicates the number of the packet. In the type and code portion, data for specifying that this packet is an echo request packet is set. The transmission time of the echo request packet is set in the option data part. Since the other party device that has received the echo request packet sends back an echo response packet including the transmission time, the voice communication device 100 compares the transmission time with the time at which the echo response packet is received. It becomes possible to know the time (round trip time) required for the response.

  The voice communication apparatus 100 starts receiving an echo response packet (also referred to as an echo packet) returned from the communication partner apparatus that has received the echo request packet at the same time as the transmission of the echo request packet is started (S102). This is performed by the echo packet receiving function 103, for example. Further, the voice communication device 100 records (saves) the transmission time and the reception time in a predetermined file or the like for each packet number (S103). That is, the voice communication apparatus 100 records the transmission timing (for example, transmission time) each time an echo request packet is transmitted. The voice communication device 100 records the reception timing (for example, reception time) every time an echo response packet is received. This recording is performed, for example, by the transmission / reception state recording function 104. This corresponds to the recording means of the present invention.

If the predetermined condition is satisfied, the voice communication device 100 ends transmission / reception of the echo request packet and the echo response packet (S104). Examples of the predetermined condition include that an end instruction has been input, that the number of transmissions has reached a preset number, or that a preset time has elapsed. When the transmission / reception of the echo request packet or the like is completed, the voice communication device 100 estimates and outputs the sound quality from the content (transmission / reception recording) recorded in the predetermined file in S103 (S105). This is performed, for example, by the transmission / reception record evaluation function 105.
(Sound quality estimation method)
Next, the sound quality estimation method (S105) will be described. This is performed by the transmission / reception record evaluation function 105 corresponding to the estimation means of the present invention.

  In the present embodiment, sound quality is estimated using a known EIF method (ITU recommendation G.113). As shown in FIG. 4, in the EIF method, the sound quality is represented by Icpif. The voice communication apparatus 100 outputs the Icpif as it is, or converts it into a display that can be easily understood by the user and outputs it. Note that the sound quality can be similarly estimated using a known E model or the like.

Icpif is calculated by the following equation 1.
(Formula 1) Icpif = Itot-A
Itot in this equation is calculated by the following equation 2.
(Formula 2) Itot = Inc + Ilr + Iq + Idte + Idd + Ie
The meaning of each element in each of these formulas is as follows.

Inc is reception side line noise, Ilr is total sound volume evaluation, Iq is quantization distortion due to A / D and D / A conversion, Idte is talker echo, Idd is delay, Ie is codec, packet loss Itot is a comprehensive evaluation of total degradation, A is a user's expectation factor,
Icpif means each expected level of user satisfaction.

  Next, specific examples of the values of the elements in these formulas will be described. For Inc, Ilr, Iq, and Idte, an optimum value or a general average value is designated. For example, each of these values is stored in advance in the storage device of the voice communication device 100, and is acquired (read) as necessary. For Idd, the average delay time is calculated based on the content recorded in the predetermined file in S103, and half of the average delay time is calculated as an estimated value of the one-way delay. Refer to the conversion table shown in FIG. Thus, Idd corresponding to the estimated value of the one-way delay is acquired (read). This corresponds to the detection means of the present invention. This Idd corresponds to a value corresponding to the round trip time of the present invention, which represents the state of the network. The conversion table is a correspondence relationship between half of the average delay time (Ta) and Idd, and is stored in the storage device of the voice communication device 100, for example.

  For Ie, referring to the conversion table shown in FIG. 6, obtain (read out) Ie corresponding to packet loss or packet loss (e.g., calculated based on echo request packet and echo response packet in voice communication apparatus 100). . This corresponds to the detection means of the present invention. This Ie corresponds to a value corresponding to the packet loss rate of the present invention, which represents the state of the network. Note that the conversion table indicates the correspondence between packet loss (%) and Ie, and is stored in the storage device of the voice communication device 100, for example. This conversion table is provided for each codec held by the voice communication apparatus 100, and the conversion table corresponding to the codec scheduled to be used is used. The psychological expectation factor A is not calculated.

As described above, the sound quality (Icpif) can be easily estimated by applying each element to the above-described Expression 1 and Expression 2.
(Sound quality output)
The voice communication apparatus 100 outputs Icpif representing the estimated sound quality as described above, or converts it into a display that can be easily understood by the user and outputs it. This is performed by the evaluation result output function 106 corresponding to the output means of the present invention.

  As an example of the latter output, for example, as shown in FIG. 7, the correspondence between the upper limit value of Icpif and the communication quality of voice conversation is stored in the storage device of the voice communication device 100 (for example, in the form of a table). , The communication quality of the voice conversation corresponding to (or the most similar to) Icpif (upper limit value) is acquired (read) from the table and output.

  As an output form of the estimated sound quality, for example, when the voice communication device 100 includes an image display device such as a liquid crystal display, information indicating the estimated sound quality is displayed on the image display device. Can be considered. For example, when the audio communication apparatus 100 includes an audio output device such as a speaker, it is conceivable that information indicating the estimated sound quality is output from the audio output device.

  As described above, according to the voice call device 100 of the present embodiment, a large-scale system is used when a call is made by sending and receiving voice packets to and from a call partner device that is a call partner side via a network. The sound quality can be estimated at a low cost without requiring a configuration.

  The above embodiments are merely examples in all respects. For this reason, this invention is limited to said embodiment and is not interpreted. That is, the present invention can be implemented in various other forms without departing from the spirit or main features thereof.

  According to the present invention, when a call is made by sending and receiving voice packets to and from a call partner apparatus that is a call partner side via a network, a large-scale system configuration is not required and the cost can be reduced from one side. Sound quality can be estimated.

It is a functional block diagram of the voice communication apparatus which is one Embodiment of this invention. It is a flowchart for demonstrating operation | movement of the audio | voice communication apparatus which is one Embodiment of this invention. It is a format example of an echo request packet (the same applies to an echo response packet described later). It is a figure for demonstrating the estimation method (EIF method) of a sound quality. It is an example of the conversion table | surface which converts a one way delay into Idd. It is an example of the conversion table which converts packet loss into Ie. It is an example of the conversion table which converts Icpif (the upper limit value) into the communication quality of voice conversation.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Voice communication apparatus 101 Input function 102 Echo packet transmission function 103 Echo packet reception function 104 Transmission / reception status recording function 105 Transmission / reception record evaluation function 106 Evaluation result output function 107 Voice codec (codec) information storage function

Claims (8)

A voice communication device for making a call by sending and receiving voice packets via a network with a call partner device that is a call partner side,
Detecting means for detecting a state of a network with the counterpart device;
Estimating means for estimating sound quality based on the state of the network detected by the detecting means;
Output means for outputting information representing the sound quality estimated by the estimating means;
A voice communication device comprising: A transmission / reception means for transmitting an echo request packet and receiving an echo response packet sent back from the partner apparatus that received the echo request packet;
Recording means for recording the transmission timing each time the echo request packet is transmitted, and recording means for recording the reception timing each time the echo response packet is received;
The detecting means detects the state of the network based on the content recorded by the recording means;
The voice communication apparatus according to claim 1. The state of the network detected by the detecting means is that a packet transmitted from the own device arrives at the partner device via the network, and a packet returned from the partner device in response to the packet arrives at the own device. It is represented by a value corresponding to the round trip time until
The voice communication apparatus according to claim 1 or 2. The state of the network detected by the detecting unit is represented by a value corresponding to a packet loss rate of a packet transmitted and received by the transmitting / receiving unit.
The voice communication apparatus according to claim 3. A method for estimating sound quality in a voice communication apparatus for performing a call by sending and receiving voice packets via a network with a call partner apparatus that is a call partner side,
A detection step of detecting a state of a network with the counterpart device;
An estimation step for estimating sound quality based on the state of the network detected by the detection step;
An output step of outputting information representing the sound quality estimated by the estimation step;
A sound quality estimation method comprising: A transmission / reception step of transmitting an echo request packet and receiving an echo response packet sent back from the partner apparatus that received the echo request packet;
A recording step of recording the transmission timing each time the echo request packet is transmitted, and recording the reception timing each time the echo response packet is received; and
The detecting step detects the state of the network based on the content recorded by the recording means;
The sound quality estimation method according to claim 5. The state of the network detected by the detecting step is that a packet transmitted from the own device arrives at the partner device via the network, and a packet returned from the partner device in response to the packet arrives at the own device. It is represented by a value corresponding to the round trip time until
The sound quality estimation method according to claim 5 or 6. The network state detected by the detection step is represented by a value corresponding to the packet loss rate of the packet transmitted and received by the transmission / reception step.
The sound quality estimation method according to claim 7.

Images