METHOD AND DEVICE FOR RECOGNITION AND METHOD AND DEVICE FOR CONSTRUCTING RECOGNITION MODEL

• US 20170148430A1
• Filed: 10/24/2016
• Published: 05/25/2017
• Est. Priority Date: 11/25/2015
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: LEE, Ho Shik
• Granted Patent: US 10,475,442 B2
• Time in Patent Office: Days
• CPC Class Codes: G10L 15/01 Assessment or evaluation of speech recognition systems
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/16 using artificial neural networks
• CPC Class Codes: G10L 2015/0635 updating or merging of old and new templates; Mean values; Weighting

Abstract

A method and a device for recognition, and a method and a device for constructing a recognition model are disclosed. A device for constructing a recognition model includes a training data inputter configured to receive additional training data, a model learner configured to train a first recognition model constructed based on basic training data to learn the additional training data, and a model constructor configured to construct a final recognition model by integrating the first recognition model with a second recognition model generated by the training of the first recognition model.

Claims

1. A device for constructing a recognition model, the device comprising:

a training data inputter configured to receive additional training data;

a model learner configured to train a first recognition model constructed based on basic training data to learn the additional training data; and

a model constructor configured to construct a final recognition model by integrating the first recognition model with a second recognition model generated by the training of the first recognition model.

2. The device of claim 1, wherein the basic training data and the additional training data comprise at least one of voice data, sound data, image data, text data, and bio-signal data.

3. The device of claim 1, wherein the first recognition model is based on a neural network.

4. The device of claim 3, wherein the model learner is configured to use either an error back-propagation algorithm or a gradient descent algorithm in training the first recognition model with the additional training data.

5. The device of claim 1, further comprising a model storage configured to store one or more first recognition models constructed based on one or more sets of basic training data in advance,

wherein, in response to the additional training data being received, the model learner is configured to acquire a first recognition model related to the received additional training data from the model storage.

6. The device of claim 1, wherein the model constructor is configured to obtain a linear sum of the first recognition model and the second recognition model.

7. The device of claim 6, wherein the model constructor is configured to select one or more weights from a preset weight candidate group, generate one or more recognition models for each weight by assigning the selected weights and performing linear summations, and determine at least one of the generated one or more recognition models as a final recognition model.

8. The device of claim 7, wherein the weights are preset based on at least one of a comparison between quantity of the basic training data and quantity of the additional training data, a purpose of additional learning, and importance of additional learning.

9. The device of claim 1, further comprising a model evaluator configured to evaluate a constructed final recognition model using evaluation data in response to the final recognition model being constructed.

10. The device of claim 9, wherein, in response to an evaluation that the final recognition model does not satisfy a predetermined standard, the model learner is configured to train the final recognition model to re-learn the additional training data.

11. A method of constructing a recognition model, the method comprising:

receiving additional training data;

training a first recognition model based on basic training data to learn the additional training data; and

constructing a final recognition model by integrating the first recognition model with a second recognition model generated based on the training of the first recognition model.

12. The method of claim 11, wherein the first recognition model is based on a neural network.

13. The method of claim 11, wherein the training of the first recognition model comprises acquiring a first recognition model related to the received additional training data from a model storage in response to receiving the additional training data.

14. The method of claim 11, wherein the constructing of the final recognition model comprises obtaining a linear sum of the first recognition model and the second recognition model.

15. The method of claim 14, wherein the constructing of the final recognition model comprises selecting one or more weights from a preset weight candidate group, generating one or more recognition models for each weight by assigning the selected weights and performing linear summations, and determining at least one of the generated one or more recognition models as a final recognition model.

16. The method of claim 11, further comprising evaluating a constructed final recognition model using evaluation data in response to the final recognition model being constructed.

17. The method of claim 16, further comprising, in response to an evaluation that the final recognition model does not satisfy a predetermined standard, training the final recognition model to re-learn the additional training data.

18. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the method of claim 11.

19. A recognition device comprising:

a processor configured to;

receive data to be recognized from a user;

determine whether to update a recognition model constructed in advance based on the received data;

in response to a determination that an update is required, train the recognition model to learn additional training data to update the recognition model; and

recognize the received data using the recognition model constructed in advance or the updated recognition model.

20. The device of claim 19, wherein the processor comprises:

a recognition data inputter configured to receive the data;

a model update determiner configured to determine whether to update the recognition model constructed in advance;

a model updater configured to train and update the recognition model; and

a data recognizer configured to recognize the received data.

21. The recognition device of claim 20, wherein the model update determiner is configured to determine whether to update the recognition model constructed in advance based on either relevance between basic training data used to construct the recognition model constructed in advance and the received data or accuracy of a recognition result obtained using the recognition model constructed in advance.

22. The recognition device of claim 20, further comprising an additional data inputter configured to receive additional training data related to the received data in response to a determination that an update is required.

23. The recognition device of claim 20, wherein the model updater is configured to perform a weighted linear summation of the recognition model constructed in advance and a recognition model generated as a result of learning of the recognition model constructed in advance and update the recognition model constructed in advance based on the result of the weighted linear summation.

24. The recognition device of claim 20, further comprising a recognition result provider configured to process a recognition result based on the purpose of the data recognition and provide the processed recognition result to the user.

25. A recognition model construction device, comprising:

a processor configured to;

train a first recognition model with additional training data to obtain a second recognition model; and

construct a third recognition model by integrating the first recognition model with the second recognition model based on a weight.

26. The device of claim 25, wherein the first recognition model is a model constructed based on basic training data, and the additional training data comprise data not included in the basic training data.

27. The device of claim 25, wherein the processor is configured to construct the third recognition model by using the weight to obtain a linear sum of the first recognition model and the second recognition model.

28. The device of claim 25, wherein the processor is further configured to select a model among a plurality of recognition models stored in a memory as the first recognition model based on a relevance of the selected model to the additional training data.