Spoken Document Retrieval using Multiple Speech Transcription Indices

• US 20090030894A1
• Filed: 06/17/2008
• Published: 01/29/2009
• Est. Priority Date: 07/23/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Mass, Yosi, Mamou, Jonathan Joseph
• Granted Patent: US 9,405,823 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3343 using phonetics
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A method and system are provided of spoken document retrieval using multiple search transcription indices. The method includes receiving a query input formed of one or more query terms and determining a type of a query term, wherein a type includes a term in a speech recognition vocabulary or a term not in a speech recognition vocabulary. One or more indices of search transcriptions are selected for searching the query term based on the type of the query term. The one or more indices are generated using different speech transcription methods. The results for the query term are scored by the one or more indices and the results of the one or more indices for the query term are merged. The results of the one or more query terms are then merged to provide the results for the query.

Claims

1. A method of spoken document retrieval using multiple search transcription indices, the method comprising:

receiving a query input formed of one or more query terms;

for each query term determining a type of the query term, wherein a type includes a term in a speech recognition vocabulary or a term not in a speech recognition vocabulary;

selecting one or more indices of search transcriptions for searching the query term based on the type of the query term;

scoring the results from the one or more indices; and

merging the results of the one or more indices for the query term.

2. The method as claimed in claim 1, including merging the results of the one or more query terms to provide results for the query.

3. The method as claimed in claim 1, wherein the one or more indices are generated using different speech transcription methods.

4. The method as claimed in claim 1, wherein the one or more indices are taken from the group of:

word indices, sub-word word-fragment indices, sub-word phonetic indices, sub-word syllable indices, word and sub-word indices using timestamps, and word and sub-word indices using offsets.

5. The method as claimed in claim 1, wherein merging of the results of the one or more indices for a query term uses a Threshold Algorithm to simultaneously scan results of different indices to aggregate retrieved document scores.

6. The method as claimed in claim 5, wherein the Threshold Algorithm for merging the results of the one or more indices for a query term using AND/OR semantics between the indices of this single term

7. The method as claimed in claim 5, wherein merging of the results of the one or more indices for a query term includes a weighted sum with weights allocated for different indices.

8. The method as claimed in claim 2, wherein merging of the results of the one or more query terms to provide results for a query uses a Threshold Algorithm to simultaneously scan results of different query terms to aggregate retrieved document scores.

9. The method as claimed in claim 8, wherein the Threshold Algorithm for merging the results of the one or more query terms for a query uses query semantics and phrase search according to the semantics in the query between query terms.

10. The method as claimed in claim 1, wherein scoring the results from the one or more indices includes one of the group of:

scoring results from a word index using weighted term frequency;

scoring results for a sub-word index using computed confusion cost of sub-word in a query term;

Boolean model scoring;

vector space scoring;

or edit distance scoring.

11. The method as claimed in claim 1, wherein for a query term formed of a term in the speech recognition vocabulary, one or more word indices and one or more sub-word indices are selected.

12. The method as claimed in claim 1, wherein for a query term formed of a term not in the speech recognition vocabulary, one or more sub-word indices are selected.

13. The method as claimed in claim 1, wherein for a query formed of at least one term in the speech recognition vocabulary and at least one term not in the speech recognition vocabulary, merging the results of query terms includes combining word and sub-word indices where the indices are combined using timestamps or offsets of the occurrence of the word or sub-word in a transcript.

14. A computer software product for spoken document retrieval using multiple search transcription indices, the product comprising a computer-readable storage medium, storing a computer in which program comprising computer-executable instructions are stored, which instructions, when read executed by a computer, perform the following steps:

receiving a query input formed of one or more query terms;

for each query term determining a type of the query term, wherein a type includes a term in a speech recognition vocabulary or a term not in a speech recognition vocabulary;

selecting one or more indices of search transcriptions for searching the query term based on the type of the query term;

scoring the results from the one or more indices; and

merging the results of the one or more indices for the query term.

15. A method of providing a service to a customer over a network for spoken document retrieval, the service comprising:

receiving a query input formed of one or more query terms;

for each query term determining a type of the query term, wherein a type includes a term in a speech recognition vocabulary or a term not in a speech recognition vocabulary;

selecting one or more indices of search transcriptions for searching the query term based on the type of the query term;

scoring the results from the one or more indices; and

merging the results of the one or more indices for the query term.

16. A search system for spoken document retrieval using multiple search transcription indices, the method comprising:

a processor;

a query input means, wherein a query is formed of one or more query terms;

means for determining a type of the query term by reference to a speech recognition vocabulary, wherein a type includes a term in a speech recognition vocabulary or a term not in a speech recognition vocabulary;

means for selecting one or more indices of search transcriptions for searching the query term based on the type of the query term;

means for scoring the results from the one or more indices; and

means for merging the results of the one or more indices for the query term.

17. The search system as claimed in claim 16, including means for merging the results of the one or more query terms to provide results for the query.

18. The search system as claimed in claim 16, wherein the one or more indices are taken from the group of:

word indices, sub-word word-fragment indices, sub-word phonetic indices, sub-word syllable indices, word and sub-word indices using timestamps, and word and sub-word indices using offsets.

19. The search system as claimed in claim 16, wherein the means for merging of the results of the one or more indices for a query term uses a Threshold Algorithm to simultaneously scan results of different indices to aggregate retrieved document scores.

20. The search system as claimed in claim 17, wherein the means for merging of the results of the one or more query terms to provide results for a query uses a Threshold Algorithm to simultaneously scan results of different query terms to aggregate retrieved document scores.