FEATURE NORMALIZATION INPUTS TO FRONT END PROCESSING FOR AUTOMATIC SPEECH RECOGNITION

• US 20150206527A1
• Filed: 07/24/2012
• Published: 07/23/2015
• Est. Priority Date: 07/24/2012
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Connolly, Dermot, Willett, Daniel
• Granted Patent: US 9,984,676 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/24 Speech recognition using non-acoustical features
• CPC Class Codes: G10L 25/24 the extracted parameters being the cepstrum

Abstract

A computer-implemented method is described for front end speech processing for automatic speech recognition. A sequence of speech features which characterize an unknown speech input is received with a computer process. A first subset of the speech features is normalized with a computer process using a first feature normalizing function. A second subset of the speech features is normalized with a computer process using a second feature normalizing function different from the first feature normalizing function. The normalized speech features in the first and second subsets are combined with a computer process to produce a sequence of mixed normalized speech features for automatic speech recognition.

Claims

1. A method comprising:

receiving, by a computing device, a sequence of speech features that characterize an unknown speech input;

normalizing selected speech features of the sequence of speech features using a plurality of different feature normalizing functions; and

combining the normalized selected speech features to produce a sequence of mixed normalized speech features for automatic speech recognition.

2. The method of claim 1, further comprising:

performing automatic speech recognition of the sequence of mixed normalized speech features.

3. The method of claim 1, wherein at least one feature normalizing function of the plurality of different feature normalizing functions is a cepstral mean normalizing function.

4. The method of claim 1, wherein at least one feature normalizing function of the plurality of different feature normalizing functions is a cepstral variance normalizing function.

5. The method of claim 1, wherein at least one feature normalizing function of the plurality of different feature normalizing functions is a maximum normalizing function.

6. The method of claim 1, wherein the unknown speech input includes speech portions and non-speech portions, and wherein one feature normalizing function of the plurality of different feature normalizing functions is used for the non-speech portions and another feature normalizing function of the plurality of different feature normalizing functions is used for the speech portions.

7. The method of claim 1, wherein the selected speech features comprise a speech energy feature and at least one feature normalizing function of the plurality of different feature normalizing functions normalizes the speech energy feature.

8. The method of claim 1, wherein the method is performed by the computing device in real time with minimum response latency.

9. The method of claim 1, wherein the normalized selected speech features are combined using linear discriminant analysis (LDA) to produce the sequence of mixed normalized speech features.

10. One or more non-transitory computer-readable media storing executable instructions that, when executed by a processor, cause a device to:

receive a sequence of speech features that characterize an unknown speech input;

normalize selected speech features of the sequence of speech features using a plurality of different feature normalizing functions; and

combine the normalized selected speech features to produce a sequence of mixed normalized speech features for automatic speech recognition.

11. The non-transitory computer-readable media of claim 10, storing executable instructions that, when executed by the processor, cause the device to:

perform automatic speech recognition of the sequence of mixed normalized speech features.

12. The non-transitory computer-readable media of claim 10, wherein at least one feature normalizing function of the plurality of different feature normalizing functions is a cepstral mean normalizing function.

13. The non-transitory computer-readable media of claim 10, wherein at least one feature normalizing function of the plurality of different feature normalizing functions is a cepstral variance normalizing function.

14. The non-transitory computer-readable media of claim 10, wherein at least one feature normalizing function of the plurality of different feature normalizing functions is a maximum normalizing function.

15. The non-transitory computer-readable media of claim 10, wherein the unknown speech input includes speech portions and non-speech portions, and wherein one feature normalizing function of the plurality of different feature normalizing functions is used for the non-speech portions and another feature normalizing function of the plurality of different feature normalizing functions is used for the speech portions.

16. The non-transitory computer-readable media of claim 10, wherein the selected speech features comprise a speech energy feature and at least one feature normalizing function of the plurality of different feature normalizing functions normalizes the speech energy feature.

18. The non-transitory computer-readable media of claim 10, wherein the normalized selected speech features are combined using linear discriminant analysis (LDA) to produce the sequence of mixed normalized speech features.

19. The non-transitory computer-readable media of claim 10, storing executable instructions that, when executed by the processor, cause the device to:

determine C0 for the sequence of speech features, wherein C0 comprises an average log-energy of a given speech frame of the sequence of speech features,

wherein normalizing the selected speech features of the sequence of speech features using the plurality of different feature normalizing functions comprises normalizing the C0 by subtracting a max C0 value calculated over the sequence of speech features, and

wherein combining the normalized selected speech features to produce the sequence of mixed normalized speech features comprises concatenating the normalized C0 with the sequence of speech features.

20. The non-transitory computer-readable media of claim 10, storing executable instructions that, when executed by the processor, cause the device to:

estimate a mean speech level of the sequence of speech features using a histogram of the sequence of speech features; and

determine a mean C0 value for the sequence of speech features using the mean speech level of the sequence of speech features,

wherein normalizing the selected speech features of the sequence of speech features using the plurality of different feature normalizing functions is performed in response to determining that the mean C0 value for the sequence of speech features is less than a target threshold.

17. (canceled)

21. A system comprising:

at least one processor; and

non-transitory memory storing computer-readable instructions that, when executed by the at least one processor, cause the system to;

receive a sequence of speech features that characterize an unknown speech input;

normalize selected speech features of the sequence of speech features using a plurality of different feature normalizing functions; and

combine the normalized selected speech features to produce a sequence of mixed normalized speech features for automatic speech recognition.