Automatic accuracy estimation for audio transcriptions

Patent metadata

Field	Value
Publication number	US-9275636-B2
Application number	US-201213463055-A
Country	US
Kind code	B2
Filing date	May 3, 2012
Priority date	May 3, 2012
Publication date	Mar 1, 2016
Grant date	Mar 1, 2016

Abstract

Embodiments of the present invention provide an approach for estimating the accuracy of a transcription of a voice recording. Specifically, in a typical embodiment, each word of a transcription of a voice recording is checked against a customer-specific dictionary and/or a common language dictionary. The number of words not found in either dictionary is determined. An accuracy number for the transcription is calculated from the number of said words not found and the total number of words in the transcription.

First claim

What is claimed is:

1. A computer-implemented method for estimating the accuracy of a transcription of a voice recording, comprising:
transcribing the voice recording into the transcription of the voice recording;
receiving the transcription of the voice recording;
providing a customer specific dictionary;
providing a dictionary of common language words;
determining a number of inaccurate words in the transcription;
determining a number of accurate words in the transcription;
determining a total number of words in the transcription;
calculating an accuracy number by dividing the number of accurate words by the total number of words;
assigning a greater weight to at least one word exceeding a predefined number of characters as compared to at least one other word below the predetermined number of characters;
retrieving, from a data structure, a set of axioms associated with at least one of the words in the transcription, wherein the set of axioms comprises a computer-parsable definition of a relationship of data to the at least one of the words; and
assigning a confidence level to each axiom of the set of axioms based on an output of a Gaussian function applied to a result of the dividing and the assigning of weight.

2. The computer-implemented method of claim 1 , wherein the step of calculating an accuracy number comprises dividing the number of inaccurate words by the total number of words.

3. The computer-implemented method of claim 1 , wherein the step of determining a number of inaccurate words comprises identifying the number of words not found in either the customer specific dictionary or the dictionary of common language words.

4. The computer-implemented method of claim 1 , wherein the step of determining a number of inaccurate words comprises identifying the number of words determined to be spelled incorrectly.

5. The computer-implemented method of claim 1 , wherein the common language words are in the English language.

6. The computer-implemented method of claim 1 , further comprising determining the length of each word in the transcription.

7. A system for estimating the accuracy of a transcription of a voice recording, comprising:
a memory medium comprising instructions;
a bus coupled to the memory medium; and
an audio transcription tool coupled to the bus that when executing the instructions causes the system to:
transcribe the voice recording into the transcription of the voice recording;
receive the transcription of the voice recording;
provide a customer specific dictionary;
provide a dictionary of common language words;
determine a number of inaccurate words in the transcription;
determine a number of accurate words in the transcription;
determine a total number of words in the transcription;
calculate an accuracy number by dividing the number of accurate words by the total number of words;
assign a greater weight to at least one word exceeding a predefined number of characters as compared to at least one other word below the predetermined number of characters;
retrieve, from a data structure, a set of axioms associated with at least one of the words in the transcription, wherein the set of axioms comprises a computer-parsable definition of a relationship of data to the at least one of the words; and
assign a confidence level to each axiom of the set of axioms based on an output of a Gaussian function applied to a result of the dividing and the assigning of weight.

8. The system of claim 7 , the memory medium further comprising instructions for causing the system to calculate the accuracy number by dividing the number of inaccurate words by the total number of words.

9. The system of claim 7 , the memory medium further comprising instructions for causing the system to determine a number of inaccurate words by identifying the number of words not found in either the customer specific dictionary or the dictionary of common language words.

10. The system of claim 7 , the memory medium further comprising instructions for causing the system to determine a number of inaccurate words by identifying the number of words determined to be spelled incorrectly.

11. The system of claim 7 , wherein the common language words are in the English language.

12. The system of claim 7 , the memory medium further comprising instructions for causing the system to determine the length of each word in the transcription.

13. A computer program product for estimating the accuracy of a transcription of a voice recording, the computer program product comprising a computer readable hardware storage device, and program instructions stored on the computer readable hardware storage device, to:
transcribe the voice recording into the transcription of the voice recording;
receive the transcription of the voice recording;
provide a customer specific dictionary;
provide a dictionary of common language words;
determine a number of inaccurate words in the transcription;
determine a number of accurate words in the transcription;
determine a total number of words in the transcription;
calculate an accuracy number by dividing the number of accurate words by the total number of words;
assign a greater weight to at least one word exceeding a predefined number of characters as compared to at least one other word below the predetermined number of characters;
retrieve, from a data structure, a set of axioms associated with at least one of the words in the transcription, wherein the set of axioms comprises a computer-parsable definition of a relationship of data to the at least one of the words; and
assign a confidence level to each axiom of the set of axioms based on an output of a Gaussian function applied to a result of the dividing and the assigning of weight.

14. The computer program product of claim 13 , the computer readable hardware storage device further comprising instructions to calculate the accuracy number by dividing the number of inaccurate words by the total number of words.

15. The computer program product of claim 13 , the computer readable hardware storage device further comprising instructions to determine a number of inaccurate words by identifying the number of words not found in either the customer specific dictionary or the dictionary of common language words.

16. The computer program product of claim 13 , the computer readable hardware storage device further comprising instructions for causing the system to determine a number of inaccurate words by identifying the number of words determined to be spelled incorrectly.

17. The computer program product of claim 13 , wherein the common language words are in the English language.

18. The computer program product of claim 13 , the computer readable hardware storage device further comprising instructions to determine the length of each word in the transcription.

19. A method for deploying a system for identifying commands for estimating the accuracy of a transcription of a voice recording, comprising:
providing a computer infrastructure being operable to:
transcribe the voice recording into the transcription of the voice recording;
receive the transcription of the voice recording;
provide a customer specific dictionary;
provide a dictionary of common language words;
determine a number of inaccurate words in the transcription;
determine a number of accurate words in the transcription;
determine a total number of words in the transcription;
calculate an accuracy number by dividing the number of accurate words by the total number of words;
assi

Assignees

• Bostick James E
• Ganci Jr John M
• Kaemmerer John P
• Trim Craig M
• IBM

Inventors

• Bostick James E
• Ganci Jr John M
• Kaemmerer John P
• Trim Craig M

Classifications

• G10L15/01 Primary

  Assessment or evaluation of speech recognition systems · CPC title

• G10L15/26

  Speech to text systems (G10L15/08 takes precedence) · CPC title

• G06F40/157

  using dictionaries or tables · CPC title

• G06F40/242

  Dictionaries · CPC title

• G06F40/189

  Automatic justification · CPC title