VOICE DATA TRANSFERRING DEVICE, TERMINAL DEVICE, VOICE DATA TRANSFERRING METHOD, AND VOICE RECOGNITION SYSTEM

• US 20120330651A1
• Filed: 06/22/2012
• Published: 12/27/2012
• Est. Priority Date: 06/22/2011
• Status: Active Grant

• Current Assignee: Clarion Co., Ltd. (Faurecia SA)
• Original Assignee: Clarion Co., Ltd. (Faurecia SA)
• Inventors: Obuchi, Yasunari, Homma, Takeshi
• Granted Patent: US 8,831,939 B2
• Time in Patent Office: Days
• US Class Current: 704/225
• CPC Class Codes: G10L 15/01 Assessment or evaluation of speech recognition systems
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

A voice data transferring device intermediates between an in-vehicle terminal and a voice recognition server. In order to check a change in voice recognition performance of the voice recognition server, the voice data transferring device performs a noise suppression processing on a voice data for evaluation in a noise suppression module; transmits the voice data for evaluation to the voice recognition server; and receives a recognition result thereof. The voice data transferring device sets a value of a noise suppression parameter used for a noise suppression processing or a value of a result integration parameter used for a processing of integrating a plurality of recognition results acquired from the voice recognition server, at an optimum value, based on the recognition result of the voice recognition server. This makes it possible to set a suitable parameter even if the voice recognition performance of the voice recognition server changes.

Claims

1. A voice data transferring device which intermediates between a terminal device and a voice recognition server, in which the terminal device:

records a voice of a user when the user is speaking;

transmits the speech voice as a voice data;

receives a recognition result of the transmitted voice data; and

outputs the recognition result to the user, and in which the voice recognition server;

receives the voice data from the terminal device;

recognizes the voice data; and

transmits the recognition result of the voice data, the voice data transferring device comprising;

a storage unit that stores therein a first parameter value used for performing a data manipulation processing on the voice data and a voice data for evaluation used for evaluating voice recognition performance of the voice recognition server;

a data processing unit that performs a data manipulation processing on the voice data for evaluation using the first parameter value, synthesizes a first voice data from the voice data for evaluation, performs a data manipulation processing on the voice data received from the terminal device using the first parameter value, and synthesizes a second voice data from the voice data received from the terminal device,

a server communication unit that transmits the first voice data to the voice recognition server, receives a first recognition result from the voice recognition server, transmits the second voice data to the voice recognition server, and receives a second recognition result from the voice recognition server;

a terminal communication unit that transmits the second recognition result of the second voice data to the terminal device; and

a parameter change unit that updates the first parameter value stored in the storage unit, based on the received first recognition result of the first voice data.

2. The voice data transferring device according to claim 1,

wherein the storage unit further stores therein history information which includes a past first voice data having been previously transmitted to the voice recognition server and a past recognition result thereof, in association with each other, and

wherein the parameter change unit references the history information, compares the first recognition result of the first voice data received by the server communication unit with the past recognition result, and, if the received first recognition result of the first voice data is different from the past recognition result, updates the first parameter.

3. The voice data transferring device according to claim 2, further comprising

a processing unit that performs:

on a voice data which have been previously recorded in an environment free from noise and whose speech contents are known, an addition processing of adding a noise data which is a data with a noise recorded therein, to the voice data, or the addition processing as well as a convolution processing of filtering the voice data according to transfer characteristics between the terminal device and a mouth of the user; and

synthesizing a voice data for evaluation.

4. The voice data transferring device according to claim 3,

wherein the processing unit:

acquires, in the addition processing, from among the voice having been recorded by the terminal device when the user was speaking the speech, a section without voice as a noise data; and

synthesizes a voice data for evaluation having a desired signal-to-noise ratio by adjusting a gain of the noise data.

5. The voice data transferring device according to claim 4,

wherein the processing unit:

acquires, in the addition processing, from among the voice having been recorded by the terminal device when the user was speaking, a voice in a section with voice; and

adjusts a gain of a voice data of the voice data for evaluation based on a power of the voice in the section with voice.

6. The voice data transferring device according to claim 1,

wherein the voice data transferring device is connected to a plurality of the voice recognition servers,

wherein the storage unit further stores therein a second parameter value which is used for integrating plural recognition results received from the plural voice recognition servers, into one recognition result,

wherein the voice data transferring device further comprises a recognition result integration unit that synthesizes one recognition result from the plural recognition results received from the plural voice recognition servers, using the second parameter value, and

wherein the terminal communication unit transmits the recognition result synthesized by the recognition result integration unit, to the terminal device.

7. The voice data transferring device according to claim 2,

wherein the voice data transferring device is connected to a plurality of the voice recognition servers,

wherein the storage unit further stores therein a second parameter value which is used for integrating plural recognition results received from the plural voice recognition servers, into one recognition result,

wherein the voice data transferring device further comprises a recognition result integration unit that synthesizes one recognition result from the plural recognition results received from the plural voice recognition servers, using the second parameter value,

wherein the parameter change unit, if the recognition result received from at least one of the voice recognition servers is different from the past recognition result in the history information, updates the second parameter value, and

wherein the terminal communication unit transmits the recognition result synthesized by the recognition result integration unit, to the terminal device.

8. A terminal device having the voice data transferring device according to claim 1, the terminal device further comprising a unit that records a voice of a user when the user is speaking, outputs the speech voice as a voice data, and outputs a recognition result of the voice data to the user.

10. A voice recognition system, comprising:

a terminal device that records a voice of a user when the user is speaking, transmits the speech voice as a voice data, receives a recognition result of the transmitted voice data, and outputs the recognition result to the user;

a voice recognition server that receives the voice data from the terminal device, recognizes the voice data, and transmits the recognition result of the voice data; and

the voice data transferring device according to claim 6,

wherein, in the voice data transferring device,

the data processing unit performs a data manipulation processing on the voice data received from the terminal device using plural different first parameter values, and synthesizes plural second voice data from the manipulated voice data,

the server communication unit transmits the plural second voice data to the voice recognition servers, and receives plural second recognition results from the voice recognition servers;

the recognition result integration unit synthesizes one recognition result from the plural recognition results received from the voice recognition servers, using the second parameter value, and

the terminal communication unit transmits the recognition result synthesized by the recognition result integration unit to the terminal device.

9. A method of transferring a voice data by a voice data transferring device which intermediates between a terminal device and a voice recognition server, in which the terminal device:

records a voice of a user when the user is speaking;

transmits the speech voice as a voice data;

receives a recognition result of the transmitted voice data; and

outputs the recognition result to the user, and in which the voice recognition server;

receives the voice data from the terminal device;

recognizes the voice data; and

transmits the recognition result of the voice data, the voice data transferring device including a storage unit that stores therein a first parameter value used for performing a data manipulation processing on the voice data and a voice data for evaluation used for evaluating voice recognition performance of the voice recognition server, the voice data transferring method comprising the steps of;

a data processing step of performing a data manipulation processing on the voice data for evaluation using the first parameter value, synthesizing a first voice data from the voice data for evaluation, performing a data manipulation processing on the voice data received from the terminal device using the first parameter value, and synthesizing a second voice data from the voice data received from the terminal device,

a server communication step of transmitting the first voice data to the voice recognition server, receiving a first recognition result from the voice recognition server, transmitting the second voice data to the voice recognition server, and receiving a second recognition result from the voice recognition server;

a terminal communication step of transmitting the second recognition result of the second voice data to the terminal device; and

a parameter change step of updating the first parameter value stored in the storage unit, based on the received first recognition result of the first voice data.