Determining navigation destination target in a situation of repeated speech recognition errors

Patent metadata

Field	Value
Publication number	US-9087515-B2
Application number	US-201113272568-A
Country	US
Kind code	B2
Filing date	Oct 13, 2011
Priority date	Oct 25, 2010
Publication date	Jul 21, 2015
Grant date	Jul 21, 2015

Abstract

A speech recognition apparatus is disclosed. The apparatus converts a speech signal into a digitalized speech data, and performs speech recognition based on the speech data. The apparatus makes a comparison between the speech data inputted the last time and the speech data inputted the time before the last time in response to a user's indication that the speech recognition results in erroneous recognition multiple times in a row. When the speech data inputted the last time is determined to substantially match the speech data inputted the time before the last time, the apparatus outputs a guidance prompting the user to utter an input target by calling it by another name.

First claim

What is claimed is:

1. A speech recognition apparatus comprising:
a speech recognition section in a processor, the speech recognition section
performs digital conversion to repeatedly convert a speech signal, which is inputted via a microphone from a user, into a speech data, the speech data being digitalized, and
performs speech recognition based on the speech data;

a comparison section in the processor, the comparison section makes a comparison between the speech data from the speech signal inputted a last time and the speech data from the speech signal inputted a time before the last time to determine whether the speech data inputted the last time substantially matches the speech data inputted the time before the last time, in response to a user's indication that the speech recognition for recognizing a name of a user-input-target facility by the speech recognition section results in erroneous recognition multiple times in a row, wherein the user-input-target facility is a specific facility that the user would like to input;
a guidance output section in the processor, the guidance output section outputs a guidance prompting the user to utter the user-input-target facility by calling the user-input-target facility by another name, when the comparison section determines that the speech data inputted the last time substantially matches the speech data inputted the time before the last time; and
a database including a dictionary in which facility names are registered,
wherein:
when the user inputs the user-input-target facility as a first speech input by uttering the name of the user-input-target facility, the speech recognition section recognizes the first speech input and refers to the dictionary of the database to determine whether a facility name coinciding with the first speech input exists in the dictionary;
when the speech recognition section determines that the facility name coinciding with the first speech input does not exist in the dictionary, the guidance output section
notifies the user that the facility name coinciding with the first speech input does not exist in the dictionary,
stores the first speech input, and
prompts the user to re-input the user-input-target facility as a second speech input;

when the speech recognition section determines that the facility name coinciding with the second speech input does not exist in the dictionary, the comparison section makes the comparison between a data of the first speech input and a data of the second speech input; and
when the comparison section determines that the data of the first speech input substantially matches the data of the second speech input, the guidance output section outputs the guidance prompting the user to utter the user-input-target facility by calling the user-input-target facility by the another name.

2. A speech recognition apparatus comprising:
a speech recognition section in a processor speech recognition section
performs digital conversion to repeatedly convert a speech signal, which is inputted via a microphone from a user, into a speech data, and
performs speech recognition based on the speech data;

a comparison section in a processor, the comparison section makes a comparison between the speech data from the speech signal inputted a last time and the speech data from the speech signal inputted a time before the last time to determine whether the speech data inputted the last time substantially matches the speech data inputted the time before the last time, in response to a user's indication that the speech recognition for recognizing a name of a user-input-target facility by the speech recognition section results in erroneous recognition multiple times in a row, wherein the user-input-target facility is a specific facility that the user would like to input;
a guidance output section in a processor, the guidance output section makes a list of names of facilities existing in an area in the vicinity of a present position and displays the list with a display device when the comparison section determines that the speech data inputted the last time substantially matches the speech data inputted the time before the last time; and
a database including a dictionary in which facility names are registered,
wherein:
when the user inputs the user-input-target facility as a first speech input by uttering the name of the user-input-target facility, the speech recognition section recognizes the first speech input and refers to the dictionary of the database to determine whether a facility name coinciding with the first speech input exists in the dictionary;
when the speech recognition section determines that the facility name coinciding with the first speech input does not exist in the dictionary, the guidance output section
notifies the user that the facility name coinciding with the first speech input does not exist in the dictionary,
stores the first speech input, and
prompts the user to re-input the user-input-target facility as a second speech input;

when the speech recognition section determines that the facility name coinciding with the second speech input does not exist in the dictionary, the comparison section makes the comparison between a data of the first speech input and a data of the second speech input; and
when the comparison section determines that the data of the first speech input substantially matches the data of the second speech input, the guidance output section makes the list of names of facilities existing in the area in the vicinity of the present position and displays the list with the display device.

3. The speech recognition apparatus according to claim 2 , wherein:
the list lists up the facility names registered in the dictionary of the database.

4. A speech recognition apparatus comprising:
a speech recognition section in a processor, the speech recognition section
performs digital conversion to repeatedly convert a speech signal, which is inputted via a microphone from a user, into a speech data, and
performs speech recognition based on the speech data;

a comparison section in the processor, the comparison section makes a comparison between the speech data from the speech signal inputted a last time and the speech data from the speech signal inputted a time before the last time to determine whether the speech data inputted the last time substantially matches the speech data inputted the time before the last time, in response to a user's indication that the speech recognition for recognizing a name of a user-input-target facility by the speech recognition section results in erroneous recognition multiple times in a row, wherein the user-input-target facility is a specific facility that the user would like to input; and
a guidance output section in the processor, the guidance output section is configured to connect with an external server via the Internet,
wherein:
when the comparison section determines that the speech data inputted the last time substantially matches the speech data inputted the time before the last time, the guidance output section
connects to the external server via the Internet makes a search by using a feature of the speech data as a search key to retrieve content, and
displays the retrieve content with a display device, and

the speech recognition apparatus uses a speech search solution enabling a search of a word with a raw speech when the search using the feature of the speech data is performed via the Internet.

5. A speech recognition apparatus for a vehicle, wherein:
a speech recognition section in a processor, the speech recognition section
performs digital conversion to repeatedly convert a speech signal, which is inputted via a microphone from a user, into a speech data, and
performs speech recognition based on the speech data;

a co

Assignees

• Tsuda Takahiro
• Denso Corp

Inventors

• Tsuda Takahiro

Classifications

• G10L15/06

  Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice (G10L15/14 takes precedence) · CPC title

• G10L15/22 Primary

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title