Pattern adapting apparatus using minimum description length criterion in pattern recognition processing and speech recognition system

• US 5,956,676 A
• Filed: 08/27/1996
• Issued: 09/21/1999
• Est. Priority Date: 08/30/1995
• Status: Expired due to Term

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Shinoda, Koichi
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): SOFOCLEOUS, MICHAEL D
• Application Number:
  US08/703,911
• Time in Patent Office: 1,120 Days
• Field of Search: 704/243, 704/255, 704/256, 704/257, 704/244, 704/252
• US Class Current: 704/244
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/144 Training of HMMs

Abstract

A pattern adapting apparatus including an input pattern forming unit, a tree structure standard pattern storing unit for storing a tree structure standard pattern including a tree structure indicative of inclusive relationships among categories and a parameter set at each node of the tree structure, a pattern matching unit for matching categories of the tree structure standard pattern with input samples of an input pattern, a tree structure standard pattern modifying unit for modifying a tree structure standard pattern based on the results of pattern matching, a node set selecting unit for calculating a description length with respect to a plurality of node sets in a tree structure pattern to select an appropriate node set, a modified standard pattern forming unit for forming a modified standard pattern by using a parameter set of a selected node set, and a standard pattern for recognition storing unit for storing a modified standard pattern.

Claims

1. A pattern adapting apparatus for adapting standard patterns made up of a plurality of categories to individual targets by learning that employs an input pattern as a set of input samples, comprising:

input pattern forming means for forming an input pattern;

tree structure standard pattern storing means for storing a tree structure standard pattern including a tree structure indicative of inclusive relationships among categories and a set of parameters at each node of the tree structure;

pattern matching means for matching the categories of said tree structure standard pattern with the input samples of said input pattern;

tree structure standard pattern modifying means for modifying said tree structure standard pattern based on the results of pattern matching by said pattern matching means;

node set selecting means for calculating a description length with respect to a plurality of node sets in said tree structure pattern to select an appropriate node set according to the calculated description length;

modified standard pattern forming means for forming a modified standard pattern by using a parameter set of the node set selected by said node set selecting means; and

standard pattern for recognition storing means for storing a modified standard pattern formed by said modified standard pattern forming means,

wherein said pattern matching means performs pattern matching by using, as a parameter, at least one of;

i) a mean vector of an output probability distribution of said standard pattern, ii) a variance of the output probability distribution of said standard pattern, and iii) a weighting factor at a state of said standard pattern.

2. The pattern adapting apparatus as set forth in claim 1, wherein a time series of frame vectors obtained by voice analysis is used as said input pattern.

3. The pattern adapting apparatus as set forth in claim 1, wherein

a time series of frame vectors obtained by voice analysis is used as said input pattern, and

a tree structure formed by using a Gaussian distribution at each state of a Hidden Markov Model whose output probability distribution is a mixed Gaussian distribution is used as said tree structure standard pattern.

12. The pattern adapting apparatus as set forth in claim 1, wherein the modified standard pattern is determined and stored in the standard pattern for recognition storing means during a voice recognition mode to be used to recognize the input pattern.

13. The pattern adapting apparatus as set forth in claim 1, wherein the modified standard pattern is determined and stored in the standard pattern for recognition storing means at a same time that the input pattern is being recognized during a voice recognition mode.

4. A pattern adapting apparatus for performing pattern recognition processing of an input voice as an input sample to specify a speaker of the input voice, comprising:

tree structure standard pattern storing means for storing a tree structure standard pattern including a tree structure indicative of inclusive relationships among categories and a set of parameters at each node of the tree structure;

pattern matching means for receiving an input pattern as a time series of frame vectors obtained by analyzing said input voice to match the categories of said tree structure standard pattern with an input sample of said input pattern;

standard pattern modifying means for modifying said tree structure standard pattern based on the results of pattern matching by said pattern matching means;

node set selecting means for calculating a description length with respect to a plurality of node sets in said tree structure pattern to select an appropriate node set according to the calculated description length;

modified standard pattern forming means for forming a modified standard pattern by using a parameter set of the node set selected by said node set selecting means; and

standard pattern for recognition storing means for storing a modified standard pattern formed by said modified standard pattern forming means,

wherein said pattern matching means performs pattern matching by using, as a parameter, at least one of;

i) a mean vector of an output probability distribution of said standard pattern, ii) a variance of the output probability distribution of said standard pattern, and iii) a weighting factor at a state of said standard pattern.

5. The pattern adapting apparatus as set forth in claim 4, wherein a tree structure formed by using a Gaussian distribution at each state of a Hidden Markov Model whose output probability distribution is a mixed Gaussian distribution is used as said tree structure standard pattern.

14. The pattern adapting apparatus as set forth in claim 4, wherein the modified standard pattern is determined and stored in the standard pattern for recognition storing means during a voice recognition mode to be used to recognize the input pattern.

15. The pattern adapting apparatus as set forth in claim 4, wherein the modified standard pattern is determined and stored in the standard pattern for recognition storing means at a same time that the input pattern is being recognized during a voice recognition mode.

6. A pattern adapting apparatus for performing pattern recognition processing of an input voice as an input sample to specify a speaker of the input voice, comprising:

tree structure standard pattern storing means for storing a tree structure standard pattern including a tree structure indicative of inclusive relationships among categories and a set of parameters at each node of the tree structure;

pattern matching means for receiving an input pattern as a time series of frame vectors obtained by analyzing said input voice to match the categories of said tree structure standard pattern with an input sample of said input pattern;

standard pattern modifying means for modifying said tree structure standard pattern based on the results of pattern matching by said pattern matching means;

node set selecting means for calculating a description length with respect to a plurality of node sets in said tree structure pattern to select an appropriate node set according to the calculated description length;

modified standard pattern forming means for forming a modified standard pattern by using a parameter set of the node set selected by said node set selecting means; and

standard pattern for recognition storing means for storing a modified standard pattern formed by said modified standard pattern forming means,

wherein a tree structure formed by using a Gaussian distribution at each state of a Hidden Markov Model whose output probability distribution is a mixed Gaussian distribution is used as said tree structure standard pattern,

and wherein said pattern matching means performs pattern matching by using a mean vector of said output probability distribution as a parameter.

7. A voice recognition system for specifying a speaker of an input voice as an input sample by performing pattern recognition processing of the input voice, including input pattern forming means for forming an input pattern as a time series of frame vectors obtained by analyzing an input voice, standard pattern storing means for storing a model of voice information source, recognition means for recognizing an input pattern based on said input pattern and said standard pattern storing means to specify a speaker of said input voice, and a pattern adapting apparatus for modifying, for an individual speaker, said model of voice information source stored in said standard pattern storing means, said pattern adapting apparatus comprising:

tree structure standard pattern storing means for storing a tree structure standard pattern including a tree structure indicative of inclusive relationships among categories and a set of parameters at each node of the tree structure;

pattern matching means for receiving an input pattern as a time series of frame vectors obtained by analyzing said input voice to match the categories of said tree structure standard pattern with an input sample of said input pattern;

standard pattern modifying means for modifying said tree structure standard pattern based on the results of pattern matching by said pattern matching means;

node set selecting means for calculating a description length with respect to a plurality of node sets in said tree structure pattern to select an appropriate node set according to the calculated description length; and

modified standard pattern forming means for forming a modified standard pattern by using a parameter set of the node set selected by said node set selecting means to store the modified standard pattern in said standard pattern storing means in place of a model of voice information source prior to the modification,

wherein said pattern matching means performs pattern matching by using, as a parameter, at least one of;

i) a mean vector of an output probability distribution of said standard pattern, ii) a variance of the output probability distribution of said standard pattern, and iii) a weighting factor at a state of said standard pattern.

8. The voice recognition system as set forth in claim 7, wherein said pattern adapting apparatus uses, as said tree structure standard pattern, a tree structure formed by using a Gaussian distribution at each state of a Hidden Markov Model whose output probability distribution is a mixed Gaussian distribution.

16. The pattern adapting apparatus as set forth in claim 7, wherein the modified standard pattern is determined and stored in the standard pattern storing means during a voice recognition mode to be used to recognize the input pattern.

17. The pattern adapting apparatus as set forth in claim 7, wherein the modified standard pattern is determined and stored in the standard pattern storing means at a same time that the input pattern is being recognized during a voice recognition mode.

9. A voice recognition system for specifying a speaker of an input voice as an input sample by performing pattern recognition processing of the input voice, including input pattern forming means for forming an input pattern as a time series of frame vectors obtained by analyzing an input voice, standard pattern storing means for storing a model of voice information source, recognition means for recognizing an input pattern based on said input pattern and said standard pattern storing means to specify a speaker of said input voice, and a pattern adapting apparatus for modifying, for an individual speaker, said model of voice information source stored in said standard pattern storing means, said pattern adapting apparatus comprising:

tree structure standard pattern storing means for storing a tree structure standard pattern including a tree structure indicative of inclusive relationships among categories and a set of parameters at each node of the tree structure;

pattern matching means for receiving an input pattern as a time series of frame vectors obtained by analyzing said input voice to match the categories of said tree structure standard pattern with an input sample of said input pattern;

standard pattern modifying means for modifying said tree structure standard pattern based on the results of pattern matching by said pattern matching means;

node set selecting means for calculating a description length with respect to a plurality of node sets in said tree structure pattern to select an appropriate node set according to the calculated description length; and

modified standard pattern forming means for forming a modified standard pattern by using a parameter set of the node set selected by said node set selecting means to store the modified standard pattern in said standard pattern storing means in place of a model of voice information source prior to the modification,

wherein said pattern adapting apparatus uses, as said tree structure standard pattern, a tree structure formed by using a Gaussian distribution at each state of a Hidden Markov Model whose output probability distribution is a mixed Gaussian distribution, and

wherein said pattern matching means of said pattern adapting apparatus performs pattern matching by using a mean vector of said output probability distribution as a parameter.

10. A pattern adapting apparatus for adapting standard patterns made up of a plurality of categories to individual targets by learning that employs an input pattern as a set of input samples, comprising:

input pattern forming means for forming an input pattern;

candidate standard pattern storing means for storing a plurality of standard patterns;

pattern matching means for matching the categories of said standard pattern with the input samples of said input pattern;

standard pattern modifying means for modifying said standard pattern based on the results of pattern matching by said pattern matching means;

description length calculating means for calculating a description length of said each modified standard pattern corresponding to said input pattern;

modified standard pattern selecting means for selecting a modified standard pattern according to a description length of said standard pattern calculated by said description length calculating means; and

standard pattern for recognition storing means for storing a modified standard pattern formed by said modified standard pattern selecting means,

wherein said pattern matching means performs pattern matching by using, as a parameter, at least one of;

i) a mean vector of an output probability distribution of said standard pattern, ii) a variance of the output probability distribution of said standard pattern, and iii) a weighting factor at a state of said standard pattern.

11. The pattern adapting apparatus as set forth in claim 10, wherein a time series of frame vectors obtained by voice analysis is used as said input pattern, and

a Hidden Markov Model whose output probability distribution is a mixed Gaussian distribution is used as said standard pattern.

18. The pattern adapting apparatus as set forth in claim 10, wherein the modified standard pattern is determined and stored in the standard pattern for recognition storing means during a voice recognition mode to be used to recognize the input pattern.

19. The pattern adapting apparatus as set forth in claim 10, wherein the modified standard pattern is determined and stored in the standard pattern for recognition storing means at a same time that the input pattern is being recognized during a voice recognition mode.