SPEECH RECOGNITION DEVICE AND METHOD, AND SEMICONDUCTOR INTEGRATED CIRCUIT DEVICE

• US 20150012275A1
• Filed: 07/07/2014
• Published: 01/08/2015
• Est. Priority Date: 07/04/2013
• Status: Active Grant

• Current Assignee: Seiko Epson Corporation (Seiko Group)
• Original Assignee: Seiko Epson Corporation (Seiko Group)
• Inventors: NONAKA, Tsutomu
• Granted Patent: US 9,190,060 B2
• Time in Patent Office: Days
• US Class Current: 704/237
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/025 Phonemes, fenemes or fenones being the recognition units
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A semiconductor integrated circuit device for speech recognition includes a scenario setting unit that receives a command designating scenario flow information and selects prescribed speech reproduction data in a speech reproduction data storage and a prescribed conversion list, in accordance with the scenario flow information, a standard pattern extraction unit that extracts a standard pattern corresponding to at least part of individual words or sentences included in the prescribed conversion list from a speech recognition database, a speech signal synthesizer that synthesizes an output speech signal, a signal processor that generates a feature pattern representing the distribution state of the frequency component of an input speech signal, and a match detector that compares the feature pattern with the standard pattern and outputs a speech recognition result.

Claims

1. A semiconductor integrated circuit device that is used in a speech recognition device that issues a question or a message to a user based on speech reproduction data and performs speech recognition processing on speech of the user, comprising:

a scenario setting unit that receives a command designating scenario flow information representing a relationship between a plurality of the speech reproduction data and a plurality of conversion lists, and, in accordance with the scenario flow information, selects prescribed speech reproduction data from among the plurality of speech reproduction data which are stored in a speech reproduction data storage, and selects a prescribed conversion list from among the plurality of conversion lists which are stored in a conversion list storage;

a standard pattern extraction unit that extracts a standard pattern corresponding to at least part of individual words or sentences included in the prescribed conversion list, from a speech recognition database containing standard patterns representing a distribution state of frequency components of a plurality of phonemes that are used in a prescribed language;

a speech signal synthesizer that synthesizes an output speech signal based on the prescribed speech reproduction data;

a signal processor that extracts the frequency component of an input speech signal by performing a Fourier-transform on the speech signal, and generates a feature pattern representing the distribution state of the frequency component of the speech signal; and

a match detector that compares the feature pattern generated from at least part of the speech signal with the standard pattern extracted from the speech recognition database, and outputs a speech recognition result.

2. The semiconductor integrated circuit device according to claim 1,

wherein the scenario setting unit, in one series of speech recognition operations, selects the prescribed speech reproduction data in accordance with the scenario flow information, from among the plurality of speech reproduction data stored in the speech reproduction data storage, and selects the prescribed conversion list set in the scenario flow information, from among the plurality of conversion lists stored in the conversion list storage.

3. The semiconductor integrated circuit device according to claim 1,

wherein after the speech signal synthesizer synthesizes an output speech signal based on first speech reproduction data selected by the selected scenario setting unit, the scenario setting unit selects second speech reproduction data set in the scenario flow information in correspondence with the speech recognition result that is output from the match detector.

4. The semiconductor integrated circuit device according to claim 3,

wherein the scenario setting unit selects the prescribed conversion list corresponding to the second speech reproduction data, in accordance with the scenario flow information, from among the plurality of conversion lists.

5. The semiconductor integrated circuit device according to claim 1,

wherein the scenario setting unit receives a command for setting or changing at least one of the speech reproduction data or at least one of the conversion lists, and sets or changes the at least one of the speech reproduction data in the speech reproduction data storage, or sets or changes the at least one of the conversion lists in the conversion list storage.

6. A speech recognition device comprising:

the semiconductor integrated circuit device according to claim 1; and

a controller that transmits the command designating scenario flow information representing the relationship between the plurality of speech reproduction data and the plurality of conversion lists to the semiconductor integrated circuit device.

7. A speech recognition device comprising:

the semiconductor integrated circuit device according to claim 2; and

a controller that transmits the command designating scenario flow information representing the relationship between the plurality of speech reproduction data and the plurality of conversion lists to the semiconductor integrated circuit device.

8. A speech recognition device comprising:

the semiconductor integrated circuit device according to claim 3; and

a controller that transmits the command designating scenario flow information representing the relationship between the plurality of speech reproduction data and the plurality of conversion lists to the semiconductor integrated circuit device.

9. A speech recognition device comprising:

the semiconductor integrated circuit device according to claim 4; and

a controller that transmits the command designating scenario flow information representing the relationship between the plurality of speech reproduction data and the plurality of conversion lists to the semiconductor integrated circuit device.

10. A speech recognition device comprising:

the semiconductor integrated circuit device according to claim 5; and

a controller that transmits the command designating scenario flow information representing the relationship between the plurality of speech reproduction data and the plurality of conversion lists to the semiconductor integrated circuit device.

11. A speech recognition method that is used in a speech recognition device that issues a question or a message to a user based on speech reproduction data and performs speech recognition processing on speech of the user, comprising:

(a) receiving a command designating scenario flow information representing a relationship between a plurality of the speech reproduction data and a plurality of conversion lists;

(b) in accordance with the scenario flow information, selecting prescribed speech reproduction data from among the plurality of speech reproduction data which are stored in a speech reproduction data storage, and selecting a prescribed conversion list from among the plurality of conversion lists which are stored in a conversion list storage;

(c) extracting a standard pattern corresponding to at least part of individual words or sentences included in the prescribed conversion list, from a speech recognition database containing standard patterns representing a distribution state of frequency components of a plurality of phonemes that are used in a prescribed language;

(d) synthesizing an output speech signal based on the prescribed speech reproduction data;

(e) extracting the frequency component of an input speech signal by performing a Fourier-transform on the speech signal, and generating a feature pattern representing the distribution state of the frequency component of the speech signal; and

(f) comparing the feature pattern generated from at least part of the speech signal with the standard pattern extracted from the speech recognition database, and outputting a speech recognition result.