Multi-pass vehicle voice recognition systems and methods

Patent metadata

Field	Value
Publication number	US-9779723-B2
Application number	US-201314409741-A
Country	US
Kind code	B2
Filing date	Jun 21, 2013
Priority date	Jun 22, 2012
Publication date	Oct 3, 2017
Grant date	Oct 3, 2017

Abstract

A voice recognition system for a vehicle includes a micro-phone for receiving speech from a user. The system further includes a memory having a partial set of commands or names for voice recognition. The memory further includes a larger set of commands or names for voice recognition. The system further includes processing electronics in communication with the microphone and the memory. The processing electronics are configured to process the received speech to obtain speech data. The processing electronics are further configured to use the obtained speech data to conduct at least two voice recognition passes. In a first pass, the speech data is compared to the partial set. In a second pass, the speech data is compared to the larger set.

First claim

What is claimed is:

1. A voice recognition system for a vehicle, comprising:
a microphone for receiving speech from a user;
a memory comprising a user specific voice recognition setting database associating at each of a plurality of users with a first set of user-specific commands or names for voice recognition and a second set of user-specific commands or names for voice recognition including at least one command or at least one name not included in the first set; and
processing electronics in communication with the microphone and the memory, wherein the processing electronics are configured to:
identify the user;
process the received speech to obtain speech data;
conduct a first speech recognition pass, the first speech recognition pass comprising:
comparing the speech data to the first set associated with the identified user to obtain a first proposed match and a first confidence of the comparison between the speech data and the first set; and
determining the first proposed match to be a first match if the first confidence exceeds a first level of confidence;

in response to the first match not being determined, conduct a second speech recognition pass, the second speech recognition pass comprising:
comparing the speech data to the second set associated with the identified user to obtain a second proposed match and a second confidence of the comparison based on the comparison between the speech data and the second set; and
determining the second proposed match to be a second match if the second confidence exceeds a second level of confidence.

2. The voice recognition system of claim 1 , wherein the memory is a part of the processing electronics.

3. The voice recognition system of claim 1 , wherein the processing electronics are configured to skip the second pass when the required first level of confidence is met or exceeded in the first pass.

4. The voice recognition system of claim 3 , wherein the processing electronics are configured to adaptively update the first level of confidence based on user feedback received at the microphone or received at a user input device.

5. The voice recognition system of claim 4 , wherein a different first level of confidence for the first pass and a different partial set of commands or names are associated with different users of the vehicle or different user devices.

6. The voice recognition system of claim 1 , wherein the first set of commands or names for voice recognition comprise the most recently used items, wherein the items comprise multiple categories of commands or names.

7. The voice recognition system of claim 1 , wherein the first set of commands or names for voice recognition comprise at least one of:
(a) a partial set of commands,
(b) a partial set of voice tags, and
(c) a set of user-specified favorite contacts or user-specified favorite phonebook names.

8. The voice recognition system of claim 1 , wherein the first set of commands or names for voice recognition comprise at least one of:
(a) a set of most frequently used commands,
(b) a set of most frequently used voice tags, and
(c) a set of most frequently used contacts or phonebook names.

9. The voice recognition system of claim 1 , wherein the processing electronics are further configured to update the first set of commands or names for voice recognition in the first pass as the frequency of use changes for the commands or names.

10. A computerized method for use by a voice recognition system for a vehicle, the method comprising:
receiving speech from a user at a microphone;
identifying the user;
loading a first set of user-specific commands or names associated with the identified user from a user specific voice recognition setting database in memory;
conducting a first voice recognition pass by comparing a representation of the received speech to the loaded first set of commands or names to obtain a first proposed match and a first confidence;
determining whether the first proposed match is a first match between the first set and the representation of the received speech if the first confidence meets or exceeds a first confidence level;
when the first proposed match is not determined to be a first match,
conducting a second voice recognition pass by comparing the representation of the received speech to a second set of user-specific at least one command or at least one name not included in the first set, the second set associated with the user in the user specific voice recognition setting database, to obtain a second proposed match and a second confidence;
determining whether the second proposed match is a second match between the second set and the representation of the received speech if the second confidence meets or exceeds a second confidence level; and
at least one of: (a) outputting an indication of the first match or the second match for user confirmation if the first match or the second match is determined, or (b) executing a vehicle function based on the first match or the second match if the first match or the second match is determined.

11. The method of claim 10 , wherein the first pass and second pass are conducted without assistance from a remote system.

12. The method of claim 10 , wherein the first set and the second set are loaded from memory local to the voice recognition system.

13. The method of claim 10 , wherein the first confidence level and the second confidence level are different.

14. The method of claim 13 , wherein the first confidence level is higher than the second confidence level.

15. The method of claim 14 , further comprising:
adaptively updating the first confidence level based on user feedback received at the microphone or received at a user input device.

16. The method of claim 15 , wherein adaptively updating comprises raising the first confidence level in response to the user feedback indicating that results of first passes were not correct.

17. The method of claim 15 , wherein adaptively updating comprises lowering the first confidence level in response to the user feedback indicating that results of the first passes were correct.

18. The method of claim 10 , further comprising:
maintaining a different first set of commands or names for each of a plurality of different users or user devices of the voice recognition system.

19. The method of claim 10 , wherein the first set of commands or names for voice recognition comprise at least one of:
(a) a set of most frequently used commands,
(b) a set of most frequently used voice tags, and
(c) a set of most frequently used contacts or phonebook names.

20. The method of claim 19 , further comprising:
updating the first set of commands or names for voice recognition in the first pass as the frequency of use changes for the commands or names.

Assignees

• Johnson Controls Tech Co
• Visteon Global Tech Inc

Inventors

• Panainte Sorin M
• Hughes David J

Classifications

• G10L2015/228

  of application context · CPC title

• G10L15/28

  Constructional details of speech recognition systems · CPC title

• G10L15/08 Primary

  Speech classification or search · CPC title

• H04R1/08

  Mouthpieces; {Microphones;} Attachments therefor · CPC title

• G10L15/32 Primary

  Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems · CPC title