SPEECH RECOGNITION DICTIONARY COMPILATION ASSISTING SYSTEM, SPEECH RECOGNITION DICTIONARY COMPILATION ASSISTING METHOD AND SPEECH RECOGNITION DICTIONARY COMPILATION ASSISTING PROGRAM

• US 20090024392A1
• Filed: 02/02/2007
• Published: 01/22/2009
• Est. Priority Date: 02/23/2006
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Koshinaka, Takafumi
• Granted Patent: US 8,719,021 B2
• Time in Patent Office: Days
• US Class Current: 704/257
• CPC Class Codes: G10L 15/06 Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice G10L15/14 takes precedence
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

A speech recognition dictionary making supporting system for efficiently making/updating a speech recognition dictionary/language model with reduced speech recognition errors by using text data available at low cost. The speech recognition dictionary making supporting system comprises a recognition dictionary storage section (105), a language model storage section (106), and a sound model storage section (107). A virtual speech recognizing section (102) creates virtual speech recognition result text data in regard to an analyzed text data created by a text analyzing section (101) with reference to a recognition dictionary, language model, and sound model, and compares the virtual speech recognition result text data with the original analyzed text data. An updating section (103) updates the recognition dictionary and language model so that the different portions in both the text data may be lessened.

Claims

1. A speech recognition dictionary compilation assisting system comprising:

a storage section storing dictionary and language models and an acoustic model;

a text analysis section applying morphological analysis to text data;

a virtual speech recognition processing section that applies the dictionary and language models to analyzed text data received from the text analysis section so as to generate virtual text data resulted from speech recognition, and extracts different point(s) between the analyzed text data and the virtual text data resulted from speech recognition; and

an update processing section that corrects at least one of the dictionary and language models in accordance to the different point(s).

2. The speech recognition dictionary compilation assisting system according to claim 1, wherein said virtual speech recognition processing section generates a sequence of feature vectors having acoustic parameters as elements from the analyzed text data and performs a virtual speech recognition process to generate a text data resulted from speech recognition.

3. The speech recognition dictionary compilation assisting system according to claim 1, wherein said storage section stores a table of distances or degrees of resemblance between elements such as states that constitute a recognition unit, and

said virtual speech recognition processing section generates a sequence of the recognition units from the analyzed text data, and searches in the dictionary and language models a string of words that has the least sum of distances or largest sum of the degrees of resemblance to generate the virtual text data resulted from speech recognition.

4. The speech recognition dictionary compilation assisting system according to claim 1, wherein said storage section stores a table of distances or degrees of resemblance between elements such as states that constitute a recognition unit, and

said virtual speech recognition processing section generates a sequence of the elements from the analyzed text data, and searches in the dictionary and language models a string of words that has the least sum of distances or largest sum of the degrees of resemblance to generate the virtual text data resulted from speech recognition.

5. The speech recognition dictionary compilation assisting system according to claim 1, wherein said update processing section adds a word that has appeared in the analyzed text data to the dictionary in accordance to the different point(s) between the analyzed text data and the virtual text data resulted from speech recognition.

6. The speech recognition dictionary compilation assisting system according to claim 1, wherein said update processing section corrects the language model such that the priority of the word or word string that has appeared in the analyzed text data becomes higher in accordance to the different point(s) between the analyzed text data and the virtual text data resulted from speech recognition.

7. The speech recognition dictionary compilation assisting system according to claim 1, wherein said update processing section corrects the language model such that the priority of the word or word string that has appeared in the virtual text data resulted from speech recognition becomes lower in accordance to the different point(s) between the analyzed text data and the virtual text data resulted from speech recognition.

8. The speech recognition dictionary compilation assisting system according to claim 6, wherein the update processing section controls the amount of changing the priority in accordance to the frequency of appearance of the word or word string in the analyzed text data and the virtual text data resulted from speech recognition.

9. A speech recognition dictionary compilation assisting method that uses a computer, comprising:

a text analysis step of, by the computer, applying morphological analysis to text data;

a step of, by the computer, generating virtual text data resulted from speech recognition in accordance to the analyzed text data that is output from the text analysis step by using a dictionary, language models and acoustic models stored in predetermined storage devices;

a step of, by the computer, comparing the analyzed text data and the virtual text data resulted from speech recognition so as to extract different point(s) therebetween; and

an updating process of, by the computer, correcting at least one of the dictionary and language models in accordance to the different point(s).

11. The speech recognition dictionary compilation assisting method according to claim 9, wherein the computer generates a sequence of feature vectors including acoustic parameters as elements from the analyzed text data, and virtually performs the speech recognition process so as to generate the virtual text data resulted from speech recognition.

12. The speech recognition dictionary compilation assisting method according to claim 9, wherein the computer generates a sequence of the recognition units from the analyzed text data in accordance to a table of distances or degrees of resemblance between elements such as states that constitute the recognition unit and searches in the dictionary and language models a string of words that has the least sum of distances or largest sum of the degrees of resemblance to generate the virtual text data resulted from speech recognition.

13. The speech recognition dictionary compilation assisting method according to claim 9, wherein the computer generates a sequence of the elements from the analyzed text data in accordance to a table of distances or degrees of resemblance between elements such as states that constitute the recognition unit and searches in the dictionary and language models a string of words that has the least sum of distances or largest sum of the degrees of resemblance to generate the virtual text data resulted from speech recognition.

14. The speech recognition dictionary compilation assisting method according to claim 9, wherein the computer adds a word that has appeared in the analyzed text data to the dictionary in accordance to a different point(s) between the analyzed text data and the virtual text data resulted from speech recognition.

15. The speech recognition dictionary compilation assisting method according claim 9, wherein the computer corrects the language model such that the priority of the word or word string that has appeared in the analyzed text data becomes higher in accordance to a different point(s) between the analyzed text data and the virtual text data resulted from speech recognition.

16. The speech recognition dictionary compilation assisting method according to claim 9, wherein the computer corrects the language model such that the priority of the word or word string that has appeared in the virtual text data resulted from speech recognition becomes lower in accordance to a different point(s) between the analyzed text data and the virtual text data resulted from speech recognition.

17. The speech recognition dictionary compilation assisting method according to claim 15, wherein the computer controls the amount of changing the priority in accordance to the frequency of occurrence of the word or word string in the analyzed text data and the virtual text data resulted from speech recognition.

10. (canceled)

18. A program that causes a computer forming a speech recognition dictionary compilation assisting system to perform:

a text analysis process that applies morphological analysis to text data;

a process that generates the virtual text data resulted from speech recognition in accordance to the analyzed text data that is output from the text analysis process by using a dictionary, language models and acoustic models stored in predetermined storage devices;

a virtual speech recognition process that compares the analyzed txt data and the virtual text data resulted from speech recognition so as to extract different point(s) therebetween; and

an updating process that corrects at least one of the dictionary and the language model in accordance to the different point(s).

19. The program according to claim 18 that causes the computer to generate a sequence of feature vectors including acoustic parameters as elements from the analyzed text data, and virtually perform the speech recognition process so as to generate the virtual text data resulted from speech recognition.

20. The program according to claim 18 that causes the computer to generate a sequence of the recognition units from the analyzed text data in accordance to a table of distances or degrees of resemblance between elements such as states that constitute a recognition unit and search in the dictionary and language models a string of words that has the least sum of distances or largest sum of the degrees of resemblance to generate the virtual text data resulted from speech recognition.

21. The program according to claim 18 that causes the computer to generate a sequence of the elements from the analyzed text data in accordance to a table of distances or degrees of resemblance between elements such as states that constitute a recognition unit and search in the dictionary and language models a string of words that has the least sum of distances or largest sum of the degrees of resemblance to generate the virtual text data resulted from speech recognition.

22. The program according to claim 18 that causes the computer to add a word that has appeared in the analyzed text data to the dictionary in accordance to the different point(s) between the analyzed text data and the virtual text data resulted from speech recognition in the updating process.

23. The program according to claim 18 that causes the computer to correct the language model such that the priority of the word or word string that has appeared in the analyzed text data becomes higher in accordance to the different point(s) between the analyzed text data and the virtual text data resulted from speech recognition in the updating process.

24. The program according to claim 18 that causes the computer to correct the language model such that the priority of the word or the word string that has appeared in the virtual text data resulted from speech recognition becomes lower in accordance to the different point(s) between the analyzed text data and the virtual text data resulted from speech recognition in the updating process.

25. The program according to claim 23 that causes the computer to control the amount of changing the priority in accordance to the frequency of occurrence of the word or word string in the analyzed text data and the virtual text data resulted from speech recognition in the updating process.