Speech recognition terminal device, speech recognition system, and speech recognition method

• US 9,349,370 B2
• Filed: 01/13/2015
• Issued: 05/24/2016
• Est. Priority Date: 01/17/2014
• Status: Active Grant

• Current Assignee: DENSO Corporation
• Original Assignee: DENSO Corporation
• Inventors: Fujisawa, Yuki, Nada, Toru
• Primary Examiner(s): Pullias, Jesse
• Application Number:
  US14/595,357
• Publication Number:
  US 20150206532A1
• Time in Patent Office: 497 Days
• Field of Search: 704231-257, 704270-275
• US Class Current: 1/1
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

A speech recognition terminal device communicable with a speech recognition device includes a request device for requesting the speech recognition device to carry out speech recognition of a acquired speech command, a determination device for determining a filler word filling a waiting time until a result of the requested speech recognition is obtained from the speech recognition device, a filler speaking device for outputting the determined filler word during the waiting time, and a response device for responding to the user based on the acquired result of the speech recognition when the result of the speech recognition is acquired from the speech recognition device.

Claims

1. A speech recognition terminal device capable of communicating with a predetermined speech recognition device that carries out speech recognition, the speech recognition terminal device comprising:

a speech acquisition device that acquires a speech command spoken by a user;

a request device that requests the speech recognition device to carry out the speech recognition of the speech command acquired by the speech acquisition device;

a determination device that determines a content of a filler word for filling a waiting time until a result of the speech recognition of the speech command requested from the request device is obtained from the speech recognition device,

a filler speaking device that outputs the content of the filler word, which is determined by the determination device, as speech information during the waiting time;

a response device that, when the result of the speech recognition is acquired from the speech recognition device, executes a process of responding to the user based on the acquired result of the speech recognition, and

a sub-recognition device that recognizes a spoken content of the speech command acquired by the speech acquisition device, wherein

the determination device creates the content of the filler word containing at least part of the spoken content of the speech command acquired by the speech acquisition device, and

the determination device creates the filler word by applying at least part of the spoken content of the speech command recognized by the sub-recognition device to a template of preset filler word.

2. The speech recognition terminal device according to claim 1, wherein:

a plurality of the templates are set on an attribute-by-attribute basis; and

the determination device creates the filler word using, out of the plurality of the templates, a template that matches the attribute of the spoken content of the speech command recognized by the sub-recognition device.

3. The speech recognition terminal device according to claim 1, wherein:

a ready-made filler word with a completed contents is preset;

the template of the filler word is preset;

when the sub-recognition device successfully recognizes the spoken content of the speech command, the determination device creates the content of the filler word containing at least part of the recognized spoken content of the speech command; and

when the sub-recognition device unsuccessfully recognizes the spoken content of the speech command, the determination device determines the ready-made filler word as an output subject.

4. A speech recognition terminal device capable of communicating with a predetermined speech recognition device that carries out speech recognition, the speech recognition terminal device comprising:

a speech acquisition device that acquires a speech command spoken by a user;

a request device that requests the speech recognition device to carry out the speech recognition of the speech command acquired by the speech acquisition device;

a determination device that determines a content of a filler word for filling a waiting time until a result of the speech recognition of the speech command requested from the request device is obtained from the speech recognition device;

a filler speaking device that outputs the content of the filler word, which is determined by the determination device, as speech information during the waiting time;

a response device that, when the result of the speech recognition is acquired from the speech recognition device, executes a process of responding to the user based on the acquired result of the speech recognition;

a sub-recognition device that recognizes a spoken content of the speech command acquired by the speech acquisition device;

a determining device that determines whether or not the content of the filler word created using the spoken content of the speech command recognized by the sub-recognition device matches the result of the speech recognition acquired from the speech recognition device; and

a notification device that notifies the user that the filler word is error when the determining device determines that the content of the filler word does not match the result of the speech recognition acquired from the speech recognition device,

wherein

the determination device creates the content of the filler word containing at least part of the spoken content of the speech command acquired by the speech acquisition device.

5. A speech recognition terminal device capable of communicating with a predetermined speech recognition device that carries out speech recognition, the speech recognition terminal device comprising:

a speech acquisition device that acquires a speech command spoken by a user;

a request device that requests the speech recognition device to carry out the speech recognition of the speech command acquired by the speech acquisition device;

a determination device that determines a content of a filler word for filling a waiting time until a result of the speech recognition of the speech command requested from the request device is obtained from the speech recognition device;

a filler speaking device that outputs the content of the filler word, which is determined by the determination device, as speech information during the waiting time; and

a response device that, when the result of the speech recognition is acquired from the speech recognition device, executes a process of responding to the user based on the acquired result of the speech recognition,

wherein;

from among a plurality of preset ready-made filler words, the determination device selects a specific ready-made filler word as an output subject, and

based on a history of usage of the plurality of ready-made filler words, the determination device selects the specific ready-made filler word as the output subject.

6. A speech recognition method for a computer system to carry out speech recognition, comprising:

performing a speech acquisition process including acquiring a speech command spoken by a user;

performing a request process including requesting a predetermined speech recognition device to carry out the speech recognition of the speech command acquired in the speech acquisition process;

performing a determination process including determining a content of a filler word for filling a waiting time until a result of the requested speech recognition of the speech command is obtained from the speech recognition device;

performing a filler speaking process including outputting the determined content of the filler word as speech information during the waiting time;

when the result of the speech recognition is acquired from the speech recognition device, performing a response process including responding to the user based on the acquired result of the speech recognition; and

performing a sub-recognition process including recognizing a spoken content of the speech command acquired in the speech acquisition process, wherein;

the determination process includes creating the content of the filler word containing at least part of the spoken content of the speech command recognized in the sub-recognition process;

the filler speaking process includes outputting the filler word determined in the determination process; and

the determination process includes creating the filler word by applying at least part of the spoken content of the speech command recognized in the sub-recognition process to a template of preset filler word.

7. The speech recognition method according to claim 6, wherein:

a plurality of the templates are set on an attribute-by-attribute basis; and

the determination process includes creating the filler word using, out of the plurality of the templates, a template that matches the attribute of the spoken content of the speech command recognized in the sub-recognition process.

8. The speech recognition method according to claim 6, wherein:

a ready-made filler word with a completed content is preset;

the template of filler word is preset; and

the determination process includes;

when the sub-recognition process successfully recognizes the spoken content, creating the content of the filler word containing at least part of the spoken content of the speech command acquired by the speech acquisition device; and

when the sub-recognition process unsuccessfully recognizes the spoken content, determining the ready-made filler word as an output subject.

9. A speech recognition method for a computer system to carry out speech recognition, comprising:

performing a speech acquisition process including acquiring a speech command spoken by a user;

performing a request process including requesting a predetermined speech recognition device to carry out the speech recognition of the speech command acquired in the speech acquisition process;

performing a determination process including determining a content of a filler word for filling a waiting time until a result of the requested speech recognition of the speech command is obtained from the speech recognition device;

performing a filler speaking process including outputting the determined content of the filler word as speech information during the waiting time;

when the result of the speech recognition is acquired from the speech recognition device, performing a response process including responding to the user based on the acquired result of the speech recognition;

performing a sub-recognition process including recognizing a spoken content of the speech command acquired in the speech acquisition process,

wherein;

the determination process includes creating the content of the filler word containing at least part of the spoken content of the speech command recognized in the sub-recognition process;

the filler speaking process includes outputting the filler word determined in the determination process;

performing a determining process including determining whether or not the content of the filler word created using the spoken content of the speech command recognized in the sub-recognition process matches the result of the speech recognition acquired from the speech recognition device; and

performing a notification process including notifying the user that the filler word is error when the determining device determines that the content of the filler word does not match the result of the speech recognition acquired from the speech recognition process.

10. A speech recognition method for a computer system to carry out speech recognition, comprising:

performing a speech acquisition process including acquiring a speech command spoken by a user;

performing a request process including requesting a predetermined speech recognition device to carry out the speech recognition of the speech command acquired in the speech acquisition process;

performing a determination process including determining a content of a filler word for filling a waiting time until a result of the requested speech recognition of the speech command is obtained from the speech recognition device;

performing a filler speaking process including outputting the determined content of the filler word as speech information during the waiting time; and

when the result of the speech recognition is acquired from the speech recognition device, performing a response process including responding to the user based on the acquired result of the speech recognition;

wherein;

the determination process includes

from among a plurality of preset ready-made filler words, selecting a specific ready-made filler word as an output subject; and

selecting the specific ready-made filler word as the output subject based on a history of usage of the plurality of ready-made filler words.