Method and apparatus to perform voice activity detection

• US 7,318,030 B2
• Filed: 09/17/2003
• Issued: 01/08/2008
• Est. Priority Date: 09/17/2003
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Guduru, Vijayakrishna Prasad
• Primary Examiner(s): ABEBE, DANIEL DEMELASH
• Application Number:
  US10/665,859
• Publication Number:
  US 20050060149A1
• Time in Patent Office: 1,574 Days
• Field of Search: 704/233, 704/243, 704/276
• US Class Current: 704/243
• CPC Class Codes: G10L 25/78 Detection of presence or absence of voice signals switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10

Abstract

A method and apparatus to perform voice detection are described.

Claims

1. A method to perform voice detection, comprising:

receiving a frame of information;

receiving an echo cancellation reference signal;

canceling echo from said frame of information;

sending said frame of information to a voice activity detector; and

determining whether said frame comprises voice information using a fuzzy logic algorithm and by measuring at least one characteristic of said frame and generating at least one frame value based on said measurements.

2. The method of claim 1, wherein said frame value is an estimate of an energy level.

3. The method of claim 2, wherein said determining further comprises:

receiving at least one frame value;

comparing said frame value with a threshold parameter;

assigning a fuzzy logic value to said frame based on said comparison; and

determining whether said frame comprises voice information based on said fuzzy logic value.

4. The method of claim 3, wherein said determining whether said frame comprises voice information based on said fuzzy logic value comprises:

comparing said fuzzy logic value with a class indicator value; and

determining whether said frame comprises voice information in accordance with said comparison of said fuzzy logic value and said class indicator value.

5. The method of claim 1, further comprising:

determining that said frame comprises voice information; and

notifying an application system that said frame comprises voice information.

6. A system, comprising:

an antenna;

a receiver connected to said antenna to receive a frame of information;

an echo canceller connected to said receiver to cancel echo; and

a voice activity detector to detect voice information in said frame using a fuzzy logic algorithm and by measuring at least one characteristic of said frame and generating at least one frame value based on said measurements.

7. The system of claim 6, further comprising a transmitter to provide an echo cancellation reference signal to said echo canceller.

8. The system of claim 6, where said voice activity detector further comprises:

an estimator to estimate energy level values; and

a voice classification module connected to said estimator to classify information for said frame.

9. The system of claim 8, wherein said voice classification module assigns fuzzy logic values to said frame based on energy level values, and determines whether said frame comprises voice information using said fuzzy logic values.

10. A voice activity detector, comprising:

an estimator to estimate energy level values;

an echo canceller connected to said estimator to cancel echo; and

a voice classification module connected to said estimator to classify information for a frame, measure at least one characteristic of said frame and generate at least one frame value based on said measurements.

11. The voice activity detector of claim 10, wherein said voice classification module assigns fuzzy logic values to said frame based on energy level values, and determines whether said frame comprises voice information using said fuzzy logic values.

12. The voice activity detector of claim 11, wherein said voice classification module compares said fuzzy logic values to class indicators, and determines whether said frame comprises voice information in accordance with said comparison.

13. An article comprising:

a computer readable storage medium;

said computer readable storage medium including stored instructions that, when executed by a processor, result in performing voice detection, by receiving a frame of information, receiving an echo cancellation reference signal, canceling echo from said frame of information, sending said frame of information to a voice activity detector; and

determining whether said frame comprises voice information using a fuzzy logic algorithm and by measuring at least one characteristic of said frame and generating at least one frame value based on said measurements.

14. The article of claim 13, wherein the stored instructions, when executed by a processor, further results in generating said at least one frame value by estimating an energy level.

15. The article of claim 14, wherein the stored instructions, when executed by a processor, further results in said determining by receiving at least one frame value, comparing said frame value with a threshold parameter, assigning a fuzzy logic value to said frame based on said comparison, and determining whether said frame comprises voice information based on said fuzzy logic value.

16. The article of claim 15, wherein the stored instructions, when executed by a processor, further results in determining whether said frame comprises voice information based on said fuzzy logic value by comparing said fuzzy logic value with a class indicator value, and determining whether said frame comprises voice information in accordance with said comparison of said fuzzy logic value and said class indicator value.

17. The article of claim 13, wherein the stored instructions, when executed by a processor, further results in determining that said frame comprises voice information, and notifying an application system that said frame comprises voice information.

18. A method to perform voice detection, comprising:

receiving a frame of information;

receiving an echo cancellation reference signal;

canceling echo from said frame of information;

sending said frame of information to a voice activity detector; and

determining whether said frame comprises voice information using at least one frame value and comparing said frame value to a spectrum of values indicating degrees of truthfulness;

said determining further comprising measuring at least one characteristic of said frame and generating at least one frame value based on said measurements.

19. The method of claim 18, wherein said frame value is an estimate of an energy level.