Method and apparatus with a personalized speech recognition model

Patent metadata

Field	Value
Publication number	US-12236941-B2
Application number	US-202117337571-A
Country	US
Kind code	B2
Filing date	Jun 3, 2021
Priority date	Dec 29, 2017
Publication date	Feb 25, 2025
Grant date	Feb 25, 2025

Abstract

A method and apparatus for personalizing a speech recognition model is disclosed. The apparatus may obtain feedback data that is a result of recognizing a first speech input of a user using a trained speech recognition model, determine whether to update the speech recognition model based on the obtained feedback data, and selectively update, dependent on the determining, the speech recognition model based on the feedback data.

First claim

What is claimed is:

1. A processor-implemented method, the method comprising:
obtaining feedback data;
determining whether to update a speech recognition model based on the obtained feedback data; and
selectively, dependent on the determining, updating the speech recognition model,
wherein the obtaining of the feedback data comprises:
receiving a guide text from a user;
receiving a first speech input corresponding to the received guide text; and
obtaining the feedback data based on the received guide text and a recognition result, from the speech recognition model, of the received first speech input.

2. The method of claim 1 , further comprising, when the speech recognition model is updated, subsequently performing recognition of input speech using the updated speech recognition model provided the input speech.

3. The method of claim 1 , wherein, when the speech recognition model is updated, the updating of the speech recognition model includes a personalizing of the speech recognition model for the user based on a performed re-training of the speech recognition model dependent on the obtained feedback data.

4. The method of claim 3 , wherein the determining of whether to update the speech recognition model comprises obtaining a temporary speech recognition model by performing the re-training of the speech recognition model based on the obtained feedback data and based on training data representing speech of multiple individuals.

5. The method of claim 3 , wherein the determining of whether to update the speech recognition model comprises obtaining a temporary speech recognition model by performing the re-training of the speech recognition model based on one or more sets of feedback data, including the obtained feedback data, accumulated since an initial point in time.

6. The method of claim 3 , wherein the determining of whether to update the speech recognition model comprises obtaining a temporary speech recognition model by performing the re-training of the speech recognition model based on one or more sets of feedback data, including the obtained feedback data, accumulated only since a point in time after an initial point in time, where the point in time, after the initial point in time, is a time or period of time in which feedback data was previously generated subsequent to feedback data generated with respect the initial point in time.

7. The method of claim 1 , wherein the method comprises obtaining a temporary speech recognition model by re-training the speech recognition model based on one or more sets of feedback data, including the obtained feedback data, and
wherein, when the speech recognition model is updated, the updating of the speech recognition model includes replacing the speech recognition model with the temporary speech recognition model.

8. The method of claim 1 , wherein the method comprises obtaining a temporary speech recognition model by re-training the speech recognition model based at least on the obtained feedback data, and
wherein the determining of whether to update the speech recognition model, with the temporary speech recognition model, is based on respective information about the speech recognition model and the temporary speech recognition model.

9. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the method of claim 1 .

10. A processor-implemented method, the method comprising:
obtaining feedback data;
determining whether to update a speech recognition model based on the obtained feedback data; and
selectively, dependent on the determining, updating the speech recognition model,
wherein the obtaining of the feedback data comprises:
receiving a first speech input of the user;
receiving, from the user, an answer text corresponding to the received first speech input; and
obtaining the feedback data based on a recognition result, from the speech recognition model, of the received first speech input and the received answer text.

11. The method of claim 10 , wherein the method comprises obtaining a temporary speech recognition model by re-training the speech recognition model based on one or more sets of feedback data, including the obtained feedback data, and
wherein, when the speech recognition model is updated, the updating of the speech recognition model includes replacing the speech recognition model with the temporary speech recognition model.

12. The method of claim 10 , wherein the method comprises obtaining a temporary speech recognition model by re-training the speech recognition model based at least on the obtained feedback data, and
wherein the determining of whether to update the speech recognition model, with the temporary speech recognition model, is based on respective information about the speech recognition model and the temporary speech recognition model.

13. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the method of claim 10 .

14. A processor-implemented method, the method comprising:
obtaining feedback data;
determining whether to update a speech recognition model based on the obtained feedback data; and
selectively, dependent on the determining, updating the speech recognition model,
wherein the obtaining of the feedback data comprises:
receiving a first speech input;
generating a guide text corresponding to a recognition, using the speech recognition model, of the received first speech input;
receiving a second speech input of the user corresponding to the generated guide text; and
obtaining the feedback data based on the generated guide text and the received second speech input.

15. The method of claim 14 , wherein the generating of the guide text includes generating the guide text to include a word or a sentence, depending on recognition results in the use of the speech recognition model, determined to be less accurate with respect to a predetermined reference.

16. The method of claim 14 , wherein, when the speech recognition model is updated, the updating of the speech recognition model includes a personalizing of the speech recognition model for the user based on a performed re-training of the speech recognition model dependent on the obtained feedback data.

17. The method of claim 16 , wherein the determining of whether to update the speech recognition model comprises obtaining a temporary speech recognition model by performing the re-training of the speech recognition model based on one or more sets of feedback data, including the obtained feedback data, accumulated since an initial point in time.

18. The method of claim 16 , wherein the determining of whether to update the speech recognition model comprises obtaining a temporary speech recognition model by performing the re-training of the speech recognition model based on one or more sets of feedback data, including the obtained feedback data, accumulated only since a point in time after an initial point in time, where the point in time, after the initial point in time, is a time or period of time in which feedback data was previously generated subsequent to feedback data generated with respect the initial point in time.

19. The method of claim 16 , wherein the determining of whether to update the speech recognition model comprises obtaining a temporary speech recognition model by performing the re-training of the speech recognition model based on the obtained feedback data and based on training data representing s

Assignees

• Samsung Electronics Co Ltd

Inventors

• Kwon Ki Soo
• Mun Minyoung
• Yoo Sanghyun

Classifications

• G10L15/32

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

• G10L2015/0635

  updating or merging of old and new templates; Mean values; Weighting · CPC title

• G10L15/07 Primary

  to the speaker · CPC title

• G10L15/22

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

• G10L2015/228

  of application context · CPC title