Speech recognition apparatus and method

• US 7,337,113 B2
• Filed: 06/13/2003
• Issued: 02/26/2008
• Est. Priority Date: 06/28/2002
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Yamamoto, Hiroki, Nakagawa, Kenichiro, Kuboyama, Hideo
• Primary Examiner(s): {hacek over (S)}mits; Talivaldis Ivars
• Assistant Examiner(s): GODBOLD, DOUGLAS
• Application Number:
  US10/460,177
• Publication Number:
  US 20040002867A1
• Time in Patent Office: 1,719 Days
• Field of Search: 704/226, 704/230, 704/231, 704/233, 704/270, 704/275
• US Class Current: 704/233
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 2015/228 of application context

Abstract

If an adaptation is made taking into consideration the noise produced in a specific operating mode of a device in a case where the noise environment changes, a decline in recognition rate is expected during operation of the device in a mode for which no adaptation is made. Accordingly, the present operating mode of the device is detected, the name of data for speech recognition corresponding to the operating mode of the device is retrieved from a table that describes data for speech recognition, the retrieved data for speech recognition corresponding to the operating mode of the device is set and speech recognition processing is executed based upon the set data.

Claims

1. A speech recognition apparatus comprising:

an input section, arranged to input speech;

a detector, arranged to detect the present operating mode of a device;

an instructor, arranged to search a memory in which has been stored information indicating a corresponding relationship between the operating mode of the device and data for speech recognition, and to issue an instruction indicative of data for speech recognition, which has been retrieved from the memory, corresponding to the operating mode of the device;

a recognition section, arranged to recognize the input speech using the data for speech recognition indicated by the instruction from said instructor;

a sensor, arranged to acquire an operating sound of the device; and

an updater, arranged to update the data for speech recognition and store the updated data in the memory, which has been indicated by the instruction from said instructor, based on the operating sound of the device,

wherein said updater performs the update at a timing when the device is turned on, or a timing when the device has performed a prescribed operation a predetermined number of times greater than one; and

wherein the device is not the speech recognition apparatus.

2. The apparatus according to claim 1, wherein the data for speech recognition includes a statistical quantity of an acoustic feature obtained by superimposing the operating sound of the device on an utterance from a person.

3. The apparatus according to claim 2, wherein said updater updates the statistical quantity of the acoustic feature.

4. The apparatus according to claim 1, wherein the data for speech recognition includes power information concerning the operating sound of the device.

5. The apparatus according to claim 4, wherein said recognition section detects speech intervals based upon the power information concerning the operating sound.

6. The apparatus according to claim 4, wherein said updater updates the power information concerning the operating sound.

7. The apparatus according to claim 1, wherein the data for speech recognition includes power spectrum information concerning the operating sound of the dance.

8. The apparatus according to claim 7, wherein said recognition section applies spectrum subtraction processing to the input speech using the power spectrum information concerning the operating sound.

9. The apparatus according to claim 8, wherein said updater updates the power spectrum information concerning the operating sound.

10. The apparatus according to claim 1, wherein said apparatus is mounted in a copier and the device is a copier.

11. A speech recognition method comprising the steps of:

inputting speech;

detecting the present operating mode of a device;

searching a memory in which has been stored information indicating a corresponding relationship between the operating mode of the device and data for speech recognition;

issuing an instruction indicative of data for speech recognition, which has been retrieved from the memory, corresponding to the operating mode of the device;

recognizing the input speech using the data for speech recognition indicated by the instruction;

acquiring an operating sound of the device; and

updating the data for speech recognition and storing the undated data in the memory, which has been indicated byte instruction, based on the operating sound of the device,

wherein the update is performed at a timing when the device is turned on, or a timing when the device has preformed a prescribed operation a predetermined number of times greater than one; and

wherein the device is not a speech recognition apparatus.

12. A computer-executable program product stored in a computer-readable storage medium comprising program code for causing a computer to perform a speech recognition method, the method comprising the steps of:

inputting speech;

detecting the present operating mode of a device;

searching a memory in which has been stored information indicating a corresponding relationship between the operating mode of the device and data for speech recognition;

issuing an instruction indicative of data for speech recognition, which has been retrieved from the memory, corresponding to the operating mode of the device;

recognizing the input speech using the data for speech recognition indicated by the instruction;

acquiring an operating sound of the device; and

updating the data for speech recognition and storing the updated data in the memory, which has been indicated by the instruction, based on the operating sound of the device,

wherein the update is performed at a timing when the device is turned on, or a timing when the device has performed a prescribed operation a predetermined number of times greater than one; and

wherein the device is not a speech recognition apparatus.