Vehicle voice recognition systems and methods

Patent metadata

Field	Value
Publication number	US-9418661-B2
Application number	US-201214116965-A
Country	US
Kind code	B2
Filing date	May 11, 2012
Priority date	May 12, 2011
Publication date	Aug 16, 2016
Grant date	Aug 16, 2016

Abstract

A voice recognition system includes a microphone for receiving speech from a user and processing electronics. The processing electronics are in communication with the microphone and are configured to use a plurality of rules to evaluate user interactions with the voice recognition system. The processing electronics automatically determine and set an expertise level in response to and based on the evaluation. The processing electronics are configured to automatically adjust at least one setting of the voice recognition system in response to the set expertise level.

First claim

What is claimed is:

1. A voice recognition system, comprising:
a microphone configured to receive speech from a user and convert the speech to an audio speech signal; and
a voice recognition engine configured to:
cause a user expertise evaluator module to detect an indication of an expertise level of the audio speech signal received from the microphone;
set an expertise level for the user in a voice recognition setting database based on the indication of the expertise level;
retrieve an expertise-to-adjustment association from an expertise database;
automatically adjust a voice tree in a voice tree database based on applying the expertise-to-adjustment association to the expertise level for the user;
process the adjusted voice tree to generate a prompt in response to the received speech; and
output the prompt to the user.

2. The voice recognition system of claim 1 , wherein the voice recognition setting database and the voice tree database are located in non-volatile memory such that the setting of the expertise level and the adjustment to the voice tree span power cycles of the voice recognition system.

3. The voice recognition system of claim 2 , wherein the adjustment to the voice tree occurs without presenting the user a graphical user interface for making the adjustment and without receiving feedback from the user regarding the particular adjustment.

4. The voice recognition system of claim 3 , wherein the the user expertise evaluator module is further configured to detect the indication of an expertise level of the audio speech signal based on at least one of:
(a) user response time, (b) command recognition accuracy, (c) command concentration, (d) number of cancel operations requested by the user, (e) number of quit operations, (f) number of exit operations, (g) number of back operations, (h) timeouts, (i) number of affirmative confirmations, (j) number of negative confirmations, (k) false starts, (l) speed of speech of user, (m) volume of speech, (n) tone of speech, (o) use of barge-in, (p) low average recognition confidence, (q) percentage of rejected recognitions, (r) percentage of completed operations, (s) percentage of aborted operations, and (t) relative usage of a voice tree branch.

5. The voice recognition system of claim 4 , wherein the user expertise evaluator module is further configured to automatically detect the indication of an expertise level in response to the microphone receiving the audio speech signal.

6. The voice recognition system of claim 5 , wherein the expertise level is selected from a set of greater than two expertise levels.

7. The voice recognition system of claim 6 , wherein the voice recognition system is configured to determine a set of voice tree adjustments associated with each expertise level.

8. The voice recognition system of claim 7 , wherein the set of setting adjustments comprises at least one of:
an end of utterance timing adjustment,
a start of utterance timing adjustment,
a graphical user interface assistance adjustment,
an audio assistance adjustment, and
a rejection threshold level adjustment.

9. The voice recognition system of claim 1 , wherein the voice recognition system is configured to continuously receive an audio speech signal via the microphone and continuously operate the voice recognition engine in response to receiving the audio speech signal, and wherein the user expertise evaluator module is further configured to detect the indication of the expertise level using a plurality of rules that are a part of an expert system and/or a knowledge-based system.

10. A computerized method for personalizing a voice recognition system, comprising:
receiving, at a microphone of the voice recognition system, speech from a user;
converting, by the microphone, the speech to an audio speech signal;
causing, by a voice recognition engine of the voice recognition system, a user expertise evaluator module to detect an indication of an expertise level of the audio speech signal received from the microphone;
setting, by the voice recognition engine, an expertise level for the user in a voice recognition database based on the indication of the expertise level;
retrieving, by the voice recognition engine, an expertise-to-adjustment association from an expertise database;
automatically adjusting, by the voice recognition engine, a voice tree in a voice tree database based on applying the expertise-to-adjustment association to the expertise level for the user;
processing, by the voice recognition engine, the adjusted voice tree to generate a prompt in response to the received speech; and
outputting, by the voice recognition engine, the prompt to the user.

11. The method of claim 10 , wherein the voice recognition setting database and the voice tree database are located in non-volatile memory such that the adjustment setting of the expertise level and the adjustment to the voice tree span power cycles of the voice recognition system.

12. The method of claim 10 , wherein the adjustment to the voice tree occurs without presenting the user a graphical user interface for making the adjustment and without receiving feedback from the user regarding the particular adjustment.

13. The method of claim 12 , wherein the user expertise evaluator module is further configured to detect the indication of an expertise level of the audio speech signal based on at least one of:
(a) user response time, (b) command recognition accuracy, (c) command concentration, (d) number of cancel operations requested by the user, (e) number of quit operations, (f) number of exit operations, (g) number of back operations, (h) timeouts, (i) number of affirmative confirmations, (j) number of negative confirmations, (k) false starts, (l) speed of speech of user, (m) volume of speech, (n) tone of speech, (o) use of barge-in, (p) low average recognition confidence, (q) percentage of rejected recognitions, (r) percentage of completed operations, (s) percentage of aborted operations, and (t) relative usage of a voice tree branch.

14. The method of claim 13 , wherein the user expertise evaluator module is further configured to automatically detect the indication of an expertise level in response to the microphone receiving the audio speech signal.

15. The method of claim 14 , further comprising:
determining a set of voice tree adjustments in response setting the expertise level.

16. The method of claim 15 , wherein the expertise level is selected from a set of greater than two expertise levels.

17. The method of claim 16 , wherein the set of setting adjustments comprises at least one of:
an end of utterance timing adjustment,
a start of utterance timing adjustment,
a graphical user interface assistance adjustment,
an audio assistance adjustment, and
a rejection threshold level adjustment;
wherein the method further comprises continuously conducting the evaluation of the user interactions and wherein the plurality of rules are a part of an expert system or a knowledge-based system.

Assignees

• Fay William
• Douthitt Brian L
• Hughes David J
• Hannum Brian
• Johnson Controls Tech Co

Inventors

• Fay William
• Douthitt Brian L
• Hughes David J
• Hannum Brian

Classifications

• G10L15/22 Primary

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

• G10L15/222

  Barge in, i.e. overridable guidance for interrupting prompts · CPC title

• G10L15/28 Primary

  Constructional details of speech recognition systems · CPC title

• H04R2499/13

  Acoustic transducers and sound field adaptation in vehicles · CPC title

• H04R1/08

  Mouthpieces; {Microphones;} Attachments therefor · CPC title