Pattern recognition

• US 20040039572A1
• Filed: 03/26/2003
• Published: 02/26/2004
• Est. Priority Date: 03/27/2002
• Status: Active Grant

• Current Assignee: Nokia Corporation
• Original Assignee: Nokia Corporation
• Inventors: Vasilache, Marcel, Kiss, Imre
• Granted Patent: US 7,269,556 B2
• Time in Patent Office: Days
• US Class Current: 704/242
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/142 Hidden Markov Models [HMMs]
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Pattern recognition, wherein a sequence of feature vectors is formed from a digitized incoming signal, the feature vectors comprising feature vector components, and at least one feature vector is compared with templates of candidate patterns by computing a distortion measure.

According to the invention, a control signal based on at least one time-dependent variable of the recognition process is formulated, and the distortion measure is computed using only a subset of the vector components of the feature vector, the subset being chosen in accordance with said control signal.

This reduces the computational complexity of the computation, as the dimensionality of the vectors involved in the computation is effectively reduced. Although such a dimension reduction decreases the computational need, it has been found not to significantly impair the classification performance.

Claims

1. A method for pattern recognition, comprising:

forming a sequence of feature vectors from a digitized incoming signal, said feature vectors comprising feature vector components,

comparing at least one feature vector with templates of candidate patterns by computing a distortion measure including distortion measure contributions,

formulating a control signal based on at least one time-dependent variable of the recognition process, and

for said at least one feature vector, computing only a subset of said distortion measure contributions using the vector components of said at least one feature vector, said subset being chosen in accordance with said control signal.

2. A method according to claim 1, wherein said time-dependent variable belongs to the group of processor load and incoming signal properties.

3. A method according to claim 1, wherein the vector components of said at least one feature vector are masked, by applying a mask, thereby reducing the number of computed distortion measure contributions in the subset.

4. A method according to claim 3, wherein said mask is dynamically computed in accordance with said control signal.

5. A method according to claim 3, wherein said mask is selected from a set of predefined masks in accordance with said control signal.

6. A method according to claim 5, wherein said predefined masks represent different reductions of the number of computed distortion measure contributions in the subset.

7. A method according to claim 3, wherein said mask at given time instances comprises all vector components in the feature vector, causing a decimation of the sequence of feature vectors.

8. A method according to claim 1, wherein specific vector components of successive feature vectors are used with a rate depending on their temporal characteristics.

9. A method according to claim 1, wherein said subset of distortion measure contributions is combined with contributions from a previously computed distortion measure.

10. A method according to claim 1, wherein said signal represents speech and the candidate patterns represent spoken utterances.

11. A method according to claim 1, wherein said templates are Gaussian mixture densities of Hidden Markov Models (HMMs).

12. A method according to claim 11, wherein said distortion measure is based on a log-likelihood.

13. A device for pattern recognition, comprising:

means for forming a sequence of feature vectors from a digitized incoming signal,

a control module adapted to formulate a control signal based on at least one time-dependent variable of the recognition process, and

means for comparing at least one feature vector with templates of candidate patterns by computing a distortion measure comprising distortion measure contributions,

wherein said comparing means are arranged to compute only a subset of the distortion measure contributions, said subset being chosen in accordance with said control signal.

14. A device according to claim 13, wherein said control module is arranged to detect a processor load of the device.

15. A device according to claim 13, further including a masking module, adapted to apply a mask to the components of said at least one feature vector, thereby reducing the number of distortion measure contributions in the subset.

16. A device according to claim 15, wherein said masking module is adapted to select said mask from a set of predefined masks in accordance with said control signal.

17. A device according to claim 15, wherein said masking module is adapted to dynamically compute said mask in accordance with said control signal.

18. A device according to claims 13, implemented as an embedded processing device, comprising

a front-end section for forming said sequence of feature vectors, and

a back-end section for providing said set of distortion measures.

19. A speech recognizer including a device according to claim 13.

20. A communication device comprising a speech recognizer according to claim 19.

21. A system for pattern recognition comprising:

means for forming a sequence of feature vectors from a digitized incoming signal,

a control module adapted to formulate a control signal based on at least one time-dependent variable of the recognition process, and

means for comparing at least one feature vector with templates of candidate patterns by computing a distortion measure comprising distortion measure contributions,

wherein said comparing means are arranged to compute only a subset of the distortion measure contributions, said subset being chosen in accordance with said control signal.

22. A computer program product, directly loadable into the memory of a computer, comprising computer program code means for performing, when executed on the computer, the steps of:

forming a sequence of feature vector from a digitized incoming signal, said feature vectors comprising feature vector components,

comparing at least one feature vector with templates of candidate patterns by computing a distortion measure including distortion measure contributions,

formulating a control signal based on at least one time-dependent variable of the recognition process, and

for said at least one feature vector, computing only a subset of said distortion measure contributions using the vector components of said at least one feature vector, said subset being chosen in accordance with said control signal.

23. A computer program product according to claim 22, stored on a computer readable medium.