JP2003111131A - Method and device for discriminating result of speech understanding test and storage medium storing discriminating program for result of speech understanding test - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten the time required for performing operation confirming tests, to make the qualities of the tests uniform, and to improve the comprehensiveness of confirming equipment when discriminating the results of speech understanding tests, which are performed for confirming the communication mode switching operation between the data communication mode and audio communication mode of a portable telephone. SOLUTION: In a method for discriminating result of speech understanding test, it is discriminated that speeches recorded in a second speech signal file while the speeches are transmitted are contained in the file when it is confirmed that the level of the file reaches a certain fixed level based on the noise level of a first speech signal file in which only noise is recorded.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice communication test result determination method and device, and a storage medium storing a voice communication test result determination program, and more particularly to a communication mode (data communication mode / voice communication mode) of a mobile phone. The present invention relates to a voice communication test result determination method and device for automatically verifying the operation confirmation regarding the switching of, and a storage medium storing a voice communication test result determination program.

[0002]

2. Description of the Related Art FIG. 7 is a diagram for explaining a conventional test method.

With a data communication card (hereinafter referred to as a PC card) or a PC connected to a mobile phone, the communication mode of the mobile phone can be normally switched from "data communication mode" to "voice communication mode". In the test to be confirmed, as shown in FIG. 7, in the conventional test method, two people each have one mobile phone, and after switching the communication mode of the mobile phone from the data communication mode to the voice communication mode, Person A who has a mobile phone of the other person utters a voice, and person B who has the mobile telephone of the receiving side hears the uttered voice, and if the voice reaches normally, it means that the communication mode switching is normal. Check.

[0004]

However, with the conventional mobile phone, PC card, and PC connected,
The test method for confirming that the communication mode of the mobile phone can be normally switched from the data communication mode to the voice communication mode needs to be performed by the number of combinations of the types of the mobile phone, PC card, PC and OS. A manual test requires a considerable number of man-hours for test operation.

Further, there is a problem that the test result must be judged manually even though the quality of the operation confirmation is required to be kept constant.

The present invention has been made in view of the above points, and a voice communication test result determination method capable of shortening the test time for operation confirmation, making the test quality uniform, and improving the comprehensiveness of confirmation devices. Another object of the present invention is to provide a storage medium storing a device and a voice communication test result determination program.

[0007]

The present invention (Claim 1) includes:
Data communication mode of mobile terminals including mobile phones and PHS,
In the voice communication test result determination method for confirming the operation of switching communication modes between voice communication modes, obtain the total sum of data in the first test result file when no audio signal is transmitted, and divide the total by the number of data. The first average value is calculated by the above, and a value larger than the first average value is set as a threshold value. Among the data in the second test result file when the audio signal is transmitted, the data that exceeds the threshold value is the first.
If the condition of is not satisfied, it is determined that the voice communication is not performed because the voice signal has not been received, and the second
If the data in the test result file that exceeds the threshold satisfies the first condition, the sum of the data that exceeds the threshold is calculated, and the total is divided by the number of data of the data that exceeds the threshold. When the second average value is obtained by the following, and the value obtained by dividing the second average value from the data of the second test result file by the first average value satisfies the second condition, the audio signal is It is determined that the received voice has been communicated.

The present invention (Claim 2) relates to a mobile phone and a PHS.
A voice communication test result judging device for confirming an operation of switching a communication mode between a data communication mode and a voice communication mode of a mobile terminal including a first, in the case of not transmitting a voice signal.
And a threshold value for setting a value larger than the first average value as a threshold value, the first average value calculating means for obtaining the first average value by dividing the total amount by the number of data Of the setting means and the data in the second test result file when the audio signal is transmitted,
If the data that exceeds the threshold does not satisfy the first condition, the voice signal has not been received, and therefore the first determination means that determines that the voice has not been communicated and the data in the second test result file If the data exceeding the threshold satisfies the first condition, the sum of the data exceeding the threshold is obtained, and the second average value is calculated by dividing the sum by the number of data of the data exceeding the threshold. When the second average value calculating means to be obtained and the value obtained by dividing the second average value from the data of the second test result file by the first average value satisfy the second condition, the audio signal Is received and voice is communicated, the second determination means is included.

Further, the present invention (Claim 3) stores a voice communication test result determination program for confirming the operation of switching the communication mode between the data communication mode and the voice communication mode of the mobile terminal including the mobile phone and PHS. A storage medium,
A first average value calculation process of obtaining a first average value by obtaining the total sum of data in the first test result file when no audio signal is transmitted, and dividing the total sum by the number of data; Among the data in the second test result file when the voice signal is transmitted, the threshold setting process of setting a larger value as the threshold and the data exceeding the threshold does not satisfy the first condition, the voice signal is The first determination process of determining that the voice was not communicated because it was not received, and if the data exceeding the threshold among the data in the second test result file satisfies the first condition, the threshold is set. A second average value calculation process of obtaining a second average value by obtaining a total sum of the data exceeding the threshold and dividing the total sum by the number of data of the data exceeding the threshold; If a value obtained by dividing the second average value by the first average value from the data of the test result file satisfies the second condition, it is determined that the voice signal is received and the voice is communicated. And the determination process of.

As described above, according to the present invention, the voice signal received by the mobile phone on the receiving side is divided into the voice receiving side and the noise only when the voice signal is not transmitted from the mobile telephone and the case where the voice signal is transmitted. Files recorded by the client terminal respectively, the data of the recorded file (test result file) are compared by the server, and only the noise is recorded (first file).
Of the test result file), the data of the file that received the audio signal (second test result file) satisfies the certain conditions, it is judged as “with audio signal”, and the device under test You can confirm that it worked properly.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, a system for conducting a voice communication test will be described.

FIG. 1 shows a system configuration according to an embodiment of the present invention. In the system shown in the figure, a client terminal 100A on the voice transmitting side and a client terminal 100B on the voice receiving side to be tested are connected to a server 200, and each client terminal 100A, B has a PC inside.
Cards 120A and 120B are attached to the mobile phones 110, respectively.
It is connected to A and B.

The server 200 is a client terminal 100.
Control A and B and judge the test result.

Next, an outline of the operation in the above configuration will be described.

FIG. 2 is a sequence chart of a voice communication test process according to the embodiment of the present invention.

The voice communication test is for confirming the operation of switching the communication mode of the mobile phone, that is, switching from the data communication mode to the voice communication mode. The voice communication test includes a test preparation stage (hereinafter,), a test implementation stage (hereinafter,), a test result determination stage (hereinafter,
) Consists of three stages. The following will be described in order.

It should be noted that the numbers in circles in FIG. 2 correspond to the numbers in circles in the following description.

Communication mode switching instruction: First, in the state where the mobile phones 110A and 110B are in the data communication mode, the server 200 sets the communication mode of the mobile phone to the voice transmitting side client terminal 100A and the voice receiving side client 100B. Instruct to switch from "data communication mode" to "voice communication mode".

The voice transmitting side client terminal 100A and the voice receiving side client terminal 100B which have received the instruction from the server 200 change the communication mode of the mobile phones 110A and B connected via the PC cards 120A and B to "data communication". Instruct to switch from "mode" to "voice communication mode". When the switching is normally performed, the voice transmitting side client terminal 100A and the voice receiving side client terminal 1
00B notifies the server 200 that the switching has been completed.

Transfer of sound reproduction WAV file: The server 200 transfers a WAV format file (hereinafter, referred to as a sound reproduction WAV file) in which a test sound is recorded, to the sound transmission side client terminal 100A.
The sound reproduction WAV file replaces the human voice in the description of the prior art. In the test, the MIC of the voice transmitting client terminal 100A, the voice receiving client terminal 100B, and the mobile phones 110A and 110B.
The / PHONE terminal is connected, the WAV file for audio reproduction is converted into an audio signal by the audio transmission side client 100A, sent to the MIC / PHONE terminal of the mobile phone 110A, and transmitted.

The sound reproduction WAV file is used to confirm that sound is normally communicated, and the file format is such that the OS of the sound transmitting side client terminal 100A can reproduce the sound. It is not limited to WAV files, but may be MP3 files or the like.

The voice transmitting client terminal 100A is
After the transfer of the WAV file for audio reproduction is completed, the server 200
Notify that the file transfer is complete.

Recording start instruction: The server 200 instructs the audio receiving side client terminal 100B to start recording.

The voice receiving client terminal 100B which has received the recording start instruction from the server 200 is the mobile phone 110.
Recording is started in a state where B is simply connected, that is, a state where no audio signal is transmitted and only noise is received (hereinafter, referred to as “first time”).

First recording end: The voice receiving side client terminal 100 B receives 1 specified by the server 200.
After the end of recording for the second time, the server 200 is notified of the end of recording.

The voice receiving client terminal 100B is
Immediately after the end of the first recording, the recording (hereinafter referred to as "the second recording")
Start). Here, the reason why the second recording is started immediately after the completion of the first recording is to reduce the influence of noise over time and improve the test quality.

Voice reproduction instruction: The server 200 receives the notification of the end of the first recording from the voice reception side client 100B, and instructs the voice transmission side client terminal 100A to reproduce voice.

The voice transmitting side client terminal 100A, which receives the voice transmitting instruction from the server 200, transmits the voice reproducing W.
The AV file is reproduced, and the transmission of the audio signal to the mobile phone 110A is started.

Audio reproduction end: The audio transmission side client terminal 100A continues transmitting the audio signal until the reproduction of the audio reproduction WAV file is completed, and after the audio reproduction is completed, notifies the server 200 of the completion of the audio reproduction.

Recording end instruction: The server 200 receives the voice reproduction end notification from the voice transmitting side client terminal 100A, and instructs the voice receiving side client terminal 100B to end the recording. The audio receiving side client terminal 100B receives the recording end instruction from the server 200 and ends the second recording.

Test Result File Transfer / Test Result Judgment: The voice receiving side client terminal 100B transfers the two test result files recorded at the first and second times to the server 200 after the end of recording.

The server 200 determines from the data of the first and second test result files whether the second test result file contains an audio signal. The method of determining the test result will be described later. If it is determined that the audio signal is included, the device under test operates normally. If it is not determined that the audio signal is included, the device under test operates normally. I didn't.

In the above, the reason for recording twice is that the first recording causes background noise (internal noise of mobile unit + internal of PC for recording) due to a difference in communication quality that changes depending on the machine environment and each call. This is because the average level of (noise + fluctuation due to telephone transmission line noise, etc.) is measured and used as a reference value in the absence of voice.

Next, the process of determining the test result in the server 200 will be described.

First, the configuration of the server 200 will be described.

FIG. 3 shows the configuration of the server according to the embodiment of the present invention.

The server 200 shown in the figure comprises a system control unit 210, a sound reproduction WAV file 220, a test result determination unit 230, a first test result file 240, and a second test result file 250.

The system control unit 210 is a WA for voice reproduction.
The V file 220 is transferred to the voice transmitting client terminal 100.
It transfers to A, sends various instructions, receives notifications from each client terminal 100, and receives test result files 240 and 250.

The test result judging section 230 reads the data of the two test result files and judges the test result.

FIG. 4 is a diagram for explaining the first test result file in the embodiment of the present invention, and FIG.
FIG. 6 is a diagram for explaining a second test result file in the embodiment of the present invention.

The first test result file 240 contains only noise-free W transferred from the voice receiving client terminal 100B to the server 200 as shown in FIG.
It is an AV file. Second test result file 250
Is the client terminal 10 on the voice receiving side as shown in FIG.
It is a WAV file including a voice signal and noise transferred from the server 0B to the server 200.

The test result files of the first test result file 240 and the second test result file 250 are files in which values (hereinafter referred to as elements) obtained by pulse code modulation (hereinafter referred to as PCM) of an audio signal are listed. . They are,
This corresponds to the values of ax and by in FIGS. 4 and 5. What is PCM
Set the audio signal level of the target analog audio signal to n
It is a method of extracting at a rate of times (this is called sampling) and digitizing (this is called quantization). The sampling value is set to 8 kHz or more in consideration of the characteristics of the telephone transmission line.

Next, in the above configuration, two WAVs are used.
A method of determining the presence / absence of an audio signal from a file will be described.

In the test result judging section 230, the total sum of the elements of the first test result file 240 is calculated, and the average is calculated by dividing the calculated total sum by the number of elements. A value larger than the calculated average value is used as the threshold. Among the elements of the second test result file 250, when the elements exceeding the threshold value satisfy a certain condition (first condition), the total sum of the elements exceeding the threshold value is calculated, and the calculated total sum exceeds the threshold value. Calculate the average value by dividing by the number of elements.

When the value obtained by dividing the calculated average value by the average value calculated from the elements of the first test result file 240 satisfies a certain condition (second condition), "there is a voice signal". To determine.

The first condition is that the second test result file 2
That is, the ratio of the number of elements exceeding the threshold value of 50 to the total number of elements exceeds 50. The recording time of the audio signal may be short due to a timing difference between the audio reproduction and the recording. However, if the audio signal exceeds a certain ratio, it is determined that the audio signal may be included. This method can be effectively applied to intermittent voice such as human voice. The second condition is that the average value calculated from the elements exceeding the threshold of the second test result file 250 exceeds a certain ratio with respect to the average value calculated from each element of the first test result file 240. That is. As a result, it is possible to determine the presence or absence of voice even if the temporal variation of noise is taken into consideration.

The method for determining the presence / absence of the audio signal from the two test result files will be described in detail below.

FIG. 6 is a flow chart of test result determination in one embodiment of the present invention.

Step 101) The average value of each element of noise included in the first test result file 240 is calculated. Since the first test result file 240 records only noise, each element changes with time depending on the presence or absence of noise (FIG. 4, a5, a6). Further, since the noise is an alternating current, there is a negative value (FIG. 4, a7).

First, in order to obtain the average value of each element of the first test result file 240, the sum of the absolute values of each element (negative values are converted to plus) is calculated. An average value (Σ | ax | / x, a value of each element: ax, the number of elements: x) is calculated by dividing the total sum by the number of all elements of the first test result file 240, and is larger than the calculated average value. Set the value as a threshold. The threshold value is a value that serves as a reference for extracting an element of the second test result file 250 in step 102 described later. Here, the value larger than the average value is set because the noise varies with time (the noise included in the second test result file 250 is higher than the noise included in the first test result file 240). This is because the numerical elements may be gathered). A value larger than the average value means a value obtained by adding a constant (α) to the average value, or a value obtained by rounding up the decimal digits of the average value (for example, 11 for 10.25).
Or 10.3). Note that the dynamic ratio may be set in consideration that the scale on the vertical axis does not have a constant interval.

Step 102) Next, a voice signal portion is extracted from each element of the second test result file 250. From the absolute value of each element of the second test result file 250, the element having a value larger than the threshold value obtained in step 101 is extracted and set as cz. In FIG. 5, b2
Is greater than or equal to the threshold, and is therefore the target of extraction.
Is not extracted because it is less than the threshold. Also, b4, b
Each element value of 8 is a negative value, but becomes an threshold value or more by taking an absolute value, and is an extraction target.

If (z / y × 100, total number of elements in second test result file: y, number of elements exceeding threshold: z) is a certain ratio (β%) or more, second test result It can be determined that the file 250 may include an audio signal, and the process proceeds to the next step 103.

If the ratio of exceeding the threshold value in each element of the second test result file 250 is not equal to or more than the constant ratio (β%), the process proceeds to step 106.

The fixed ratio (β%) can be changed depending on the type of mobile terminal, PC card, client terminal, and OS, and even if the same ratio (for example, 10%) is applied in all tests. I don't care.

Step 103) Next, the second test result file 250 extracted in step 102 described above.
The average value is calculated for the element (cz) of. The average value is obtained by calculating the sum of absolute values of elements exceeding the threshold value and dividing the sum by the number of elements exceeding the threshold value (Σ | cz | / z, the value of each element exceeding the threshold value: cz , The number of elements exceeding the threshold value: z) can be calculated.

Step 104) Among the respective elements of the second test result file 250 for the average value of the respective elements of the first test result file 240 calculated from the respective elements of the first test result file 240, the elements exceeding the threshold value The average value ratio ((Σ | cz | / z) / (Σ | ax | / x)) is calculated, and it is determined whether or not the average value ratio exceeds a certain ratio (γ times). Step 105) If the ratio of this average value exceeds a certain ratio (γ times), it can be determined that the second test result file 250 contains an audio signal.

Step 106) If the ratio of the average values does not exceed a certain ratio (γ times), and in Step 102, the ratio of exceeding the threshold value of each element of the second test result file 250 is constant. If it is less than the ratio (β%), it can be determined that the second test result file 250 does not include an audio signal.

Here, the constant ratio (γ times) means that the ratio obtained from the average value can be changed depending on the types of the mobile terminal, the PC card, the client terminal and the OS, and in all the tests. Same ratio (eg 2.5
Double) may be applied.

In the above embodiment, the description has been given based on the configuration of the server 200 shown in FIG. 3, but the components of the server shown in FIG. Connected hard disk device, floppy disk, CD
-The present invention can be easily realized by storing it in a portable storage medium such as a ROM and installing it when carrying out the present invention.

The present invention is not limited to the above embodiment, but various modifications and applications are possible within the scope of the claims.

[0061]

In the prior art, the test for confirming that the communication mode of the mobile phone can be normally switched from the data communication mode to the voice communication mode with the mobile phone, the PC card and the PC connected is carried out by two mobile phones. Although switching from a data signal to a voice signal is manually performed by a human voice using a telephone, and considerable man-hours are required, as described above, according to the present invention, in the voice communication test, the Voice signal is not transmitted from the telephone. The voice signal received by the receiving mobile phone is recorded by the client terminal on the voice receiving side separately for noise only and when the voice signal is transmitted, and the recorded file is stored on the server. The data of the file is compared, and it is determined that there is an audio signal when the data of the file that received the audio signal meets the certain conditions for the data of the file that recorded only noise , In which equipment under test is able to see that the operation is successful.

As a result, it is possible to automatically carry out a voice communication test which can be conventionally confirmed only manually, and the required number of combinations of voices depending on the number of types of mobile phones, PC cards, PCs and OSs. In the communication test, it is possible to shorten the test time for operation confirmation, make the test quality uniform, and improve the comprehensiveness of confirmation devices.

[Brief description of drawings]

FIG. 1 is a system configuration diagram in an embodiment of the present invention.

FIG. 2 is a sequence chart of voice communication test processing according to the embodiment of the present invention.

FIG. 3 is a configuration diagram of a server according to the embodiment of the present invention.

FIG. 4 is a diagram for explaining a first test result file in the embodiment of the present invention.

FIG. 5 is a diagram for explaining a second test result file in the embodiment of the present invention.

FIG. 6 is a flowchart of test result determination according to the embodiment of the present invention.

FIG. 7 is a diagram for explaining a conventional test method.

[Explanation of symbols]

100 client terminals 110 mobile phone 120 PC card 200 servers 210 system control unit 220 WAV file for audio playback 230 Test Result Judgment Section 240 First test result file 250 Second test result file

Claims (3)

[Claims] 1. A voice communication test result determining method for confirming an operation of switching a communication mode between a data communication mode and a voice communication mode of a mobile terminal including a mobile phone and a PHS, wherein a first method for transmitting no voice signal is provided. The sum of the data in the test result file is obtained, the first sum is obtained by dividing the sum by the number of data, and a value larger than the first mean value is set as a threshold value, and when the voice signal is transmitted, Of the data in the test result file of No. 2, when the data exceeding the threshold does not satisfy the first condition, it is determined that the voice signal has not been received, and thus the voice is not communicated. Of the data in the result file, if the data exceeding the threshold value satisfies the first condition, the total sum of the data exceeding the threshold value is calculated, To obtain a second average value by dividing by the number of data of data exceeding the threshold value, and a value obtained by dividing the second average value from the data of the second test result file by the first average value. However, when the second condition is satisfied, it is determined that the voice signal is received and the voice is communicated. 2. A voice communication test result determining device for confirming an operation of switching a communication mode between a data communication mode and a voice communication mode of a mobile terminal including a mobile phone and a PHS, which is the first device when a voice signal is not transmitted. A first average value calculating means for obtaining a first average value by obtaining a total sum of data in the test result file No. 1 and dividing the total sum by the number of data, and a value larger than the first average value is set as a threshold value. Among the data in the second test result file when the audio signal is transmitted, if the data exceeding the threshold does not satisfy the first condition, the audio signal has not been received. Out of the data in the second test result file, the data that exceeds the threshold among the data in the second test result file determines the first condition. In the case of doing so, a second average value calculating means for obtaining a second average value by obtaining a total sum of data exceeding the threshold value and dividing the total sum by the number of data pieces of the data exceeding the threshold value, When the value obtained by dividing the second average value from the data of the second test result file by the first average value satisfies the second condition, the voice signal is received and the voice is communicated. A voice communication test result determination device, comprising: 3. A storage medium storing a voice communication test result determination program for confirming an operation of switching a communication mode between a data communication mode and a voice communication mode of a mobile terminal including a mobile phone and a PHS. A first average value calculation process of obtaining a first average value by obtaining a total sum of data in the first test result file when not transmitting, and dividing the total sum by the number of data; A threshold value setting process of setting a value as a threshold value, and of the data in the second test result file when the audio signal is transmitted, if the data exceeding the threshold value does not satisfy the first condition, the audio signal is received. Of the data in the second test result file, the first determination process of determining that the voice has not been communicated, When the data exceeding the value satisfies the first condition, the sum of the data exceeding the threshold is obtained, and the sum is divided by the number of data of the data satisfying the first condition to obtain the second value. And a second average value calculation process for obtaining an average value of, and a value obtained by dividing the second average value from the data of the second test result file by the first average value satisfies the second condition. Storage medium that stores a voice communication test result determination program, wherein the storage medium stores a voice communication test result determination program, the second determination process determining that the voice signal is received, and the voice is communicated.