Control using multiple speech receptors in an in-vehicle speech recognition system

• US 7,016,836 B1
• Filed: 08/30/2000
• Issued: 03/21/2006
• Est. Priority Date: 08/31/1999
• Status: Expired due to Fees

• Current Assignee: Pioneer Corporation
• Original Assignee: Pioneer Corporation
• Inventors: Yoda, Shoutarou
• Primary Examiner(s): Chawan, Vijay B.
• Application Number:
  US09/651,058
• Time in Patent Office: 2,029 Days
• Field of Search: 704270-275, 704/270.1, 704/233, 704/248, 704/227, 704/234, 704/232, 704/240, 704/231, 379/206.01, 455/563, 455/575.9
• 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 2021/02166 Microphone arrays; Beamforming

Abstract

Disclosed are a speech recognition system which comprises the following components, and a speech recognition method for this speech recognition system. The speech recognition system comprises a plurality of voice pickup sections for picking up uttered voices, a determination section for determining a speech signal suitable for speech recognition from speech signals output from the plurality of voice pickup sections, and a speech recognizer for performing speech recognition based on the speech signal determined by the determination section.

Claims

1. A speech recognition system comprising:

a plurality of voice pickup means for picking up uttered voices;

determination means for determining a speech signal suitable for speech recognition from speech signals output from said plurality of voice pickup means; and

speech recognition means for performing speech recognition based on said speech signal determined by said determination means,

wherein said determination means acquires an average S/N value and average voice power of each of said speech signals output from said plurality of voice pickup means and selects said speech signal whose average S/N value and average voice power are greater than respective predetermined threshold values as said speech signal suitable for speech recognition, and

wherein said determination means determines a candidate order of those speech signals whose average S/N values and average voice powers are greater than said respective predetermined threshold values and which are candidates for said speech signal suitable for speech recognition, in accordance with said average S/N values and average voice powers; and

said speech recognition means sequentially executes speech recognition on said candidates in accordance with said candidate order from a highest candidate to a lower one.

2. A speech recognition system comprising:

a plurality of voice pickup sections for picking up uttered voices;

a determination section for determining a speech signal suitable for speech recognition from speech signals output from said plurality of voice pickup sections; and

a speech recognizer for performing speech recognition based on said speech signal determined by said determination section,

wherein said determination section acquires an average S/N value and average voice power of each of said speech signals output from said plurality of voice pickup sections and selects said speech signal whose average S/N value and average voice power are greater than respective predetermined threshold values as said speech signal suitable for speech recognition, and

wherein said determination section determines a candidate order of those speech signals whose average S/N values and average voice powers are greater than said respective predetermined threshold values and which are candidates for said speech signal suitable for speech recognition, in accordance with said average S/N values and average voice powers; and

said speech recognizer sequentially executes speech recognition on said candidates in accordance with said candidate order from a highest candidate to a lower one.

3. A speech recognition method for a speech recognition system having a plurality of voice pickup means for picking up voices, comprising:

a voice pickup step of picking up uttered voices using said plurality of voice pickup means;

a determination step of determining a speech signal suitable for speech recognition from speech signals output from said plurality of voice pickup means; and

a speech recognition step of performing speech recognition based on said speech signal determined by said determination step,

wherein said determination step includes a step of acquiring an average S/N value and average voice power of each of said speech signals output from said plurality of voice pickup means and selecting said speech signal whose average S/N value and average voice power are greater than respective predetermined threshold values as said speech signal suitable for speech recognition,

wherein said determination step further includes a step of determining a candidate order of those speech signals whose average S/N values and average voice powers are greater than said respective predetermined threshold values and which are candidates for said speech signal suitable for speech recognition, in accordance with said average S/N values and average voice powers; and

said speech recognition step sequentially executes speech recognition on said candidates in accordance with said candidate order from a highest candidate to a lower one.