Target specific data filter to speed processing

• US 7,831,424 B2
• Filed: 04/02/2008
• Issued: 11/09/2010
• Est. Priority Date: 07/07/2006
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Verscheure, Olivier, Huerta, Juan M., Ramaswamy, Ganesh N., Chaudhari, Upendra V.
• Primary Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US12/060,903
• Publication Number:
  US 20090043579A1
• Time in Patent Office: 951 Days
• Field of Search: None
• US Class Current: 704/247
• CPC Class Codes: G10L 17/02 Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction

Abstract

A method is presented which reduces data flow and thereby increases processing capacity while preserving a high level of accuracy in a distributed speech processing environment for speaker detection. The method and system of the present invention includes filtering out data based on a target speaker specific subset of labels using data filters. The method preserves accuracy and passes only a fraction of the data by optimizing target specific performance measures. Therefore, a high level of speaker recognition accuracy is maintained while utilizing existing processing capabilities.

Claims

1. A method for reducing the amount of speech data for detecting desired speakers, comprising:

receiving a plurality of speech data from a plurality of speakers using a computer program stored on a computer;

labeling the speech data using the computer program;

priority ranking the labeled speech data using a speaker dependent filter for each of the speakers using the computer program;

generating the priority ranking by prioritizing the labeled speech data according to a performance criterion for each of the plurality of speakers using the computer program;

analyzing the labeled speech data according to the priority ranking of the labeled speech data for a specific speaker;

filtering out specified lower priority labeled speech data such that less of the speech data is analyzed; and

adjusting the step of filtering out specified lower priority labeled speech data to attain a specified speech data throughput.

2. The method of claim 1, wherein the speakers are divided into groups of speakers and the filters are generated and used for each group of speakers.

3. The method of claim 2, wherein the labels for the groups of speakers are priority ranked.

4. The method of claim 1, wherein the data is pre-labeled.

5. The method of claim 1, further comprising:

creating a group of filtered speech data for recognition analysis using the speaker dependent filter, and the group of filtered speech data being smaller than the plurality of speech data; and

analyzing the filtered group of speech data for recognition.

6. A speaker recognition system using a computer, which comprises:

a plurality of speech data from a plurality of speakers stored on a computer;

a labeler element for analyzing the speech data and labeling the data;

a speech data filter including a filter element for arranging the labels in a priority rank and the speech data filter designed according to a performance criterion for each of the plurality of speakers;

a recognition element for analyzing the labeled data in accordance with the priority rank as determined by the speech data filter for filtering out specified lower priority labeled speech data to reduce speech data analyzed; and

a specified speech data throughput being attained by filtering out specified lower priority labeled speech data.

7. The system of claim 6, wherein the labels are phonetic.

8. A computer program product comprising a computer having a computer program, the computer executing the steps of the computer program for reducing the amount of speech data when detecting desired speakers, the program steps comprising:

receiving a plurality of speech data from a plurality of speakers using the computer program stored on the computer;

labeling the speech data using the computer;

priority ranking the labeled speech data using a speaker dependent filter for each of the speakers using the computer;

generating the priority ranking by prioritizing the labeled speech data according to a performance criterion for each of the plurality of speakers using the computer;

analyzing the labeled speech data according to the priority ranking of the labeled speech data for a specific speaker;

filtering out specified lower priority labeled speech data such that less of the speech data is analyzed; and

adjusting the step of filtering out specified lower priority labeled speech data to attain a specified speech data throughput.