METHOD AND APPARATUS FOR VOICED SPEECH DETECTION

• US 20180261239A1
• Filed: 05/10/2018
• Published: 09/13/2018
• Est. Priority Date: 11/19/2015
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: FALK, Tommy, POBLOTH, Harald, KARLSSON, Erlendur
• Granted Patent: US 10,825,472 B2
• Time in Patent Office: Days
• CPC Class Codes: G10L 25/21 the extracted parameters being power information
• CPC Class Codes: G10L 25/84 for discriminating voice from noise
• CPC Class Codes: G10L 25/90 Pitch determination of speech signals
• CPC Class Codes: G10L 25/93 Discriminating between voiced and unvoiced parts of speech signals G10L25/90 takes precedence

Abstract

Detecting voiced speech in an audio signal. A method comprises calculating an autocorrelation function (ACF) of a portion of an input audio signal and detecting a highest peak of said autocorrelation function within a determined range. A peak width and a peak height of said detected highest peak are determined and based on the peak width and the peak height it is decided whether a segment of an input audio signal comprises voiced speech.

Claims

1. A method for audio signal processing, the method comprising:

calculating a correlation function of a portion of an input audio signal;

detecting a highest peak of said correlation function;

determining a peak width of said peak;

determining a peak height of said peak;

comparing the peak width with a first threshold;

comparing the peak height with a second threshold; and

deciding based on the peak width and the peak height whether a segment of the input audio signal comprises voiced speech.

2. The method of claim 1, wherein the segment of an input audio signal is decided to comprise voiced speech as a result of determining that the peak height exceeds the first threshold and the peak width is less than the second threshold.

3. The method of claim 1, wherein the segment of the input audio signal is decided not to comprise voiced speech as a result of determining that the peak height exceeds the first threshold and the peak width exceeds the second threshold.

4. The method of claim 3, wherein the second threshold is set to a constant value.

5. The method of claim 3, wherein the second threshold is dynamically set depending on a previously detected pitch.

6. The method of claim 3, wherein the second threshold is dynamically set depending on pitch of said detected highest peak.

7. The method of claim 1, wherein the peak width is determined by:

calculating number of bins upwards from the middle of the peak before the correlation curve falls below a fall-off threshold;

calculating number of bins downwards from the middle of the peak before the correlation curve falls below said fall-off threshold; and

adding the numbers of calculated bins to indicate the peak width.

8. A computer program product comprising a non-transitory computer readable medium storing a computer program comprising computer readable code units which when run on an apparatus causes the apparatus to perform the method of claim 1.

9. An apparatus comprising:

a processor, and

a memory storing instructions that, when executed by the processor, cause the apparatus to;

calculate a correlation function of a portion of an input audio signal;

detect a highest peak of said correlation function;

determine a peak width of said peak;

determine a peak height of said peak;

compare the peak width with a first threshold;

compare the peak height with a second threshold; and

decide based on the peak width and the peak height whether a segment of the input audio signal comprises voiced speech.

10. The apparatus of claim 9, wherein the apparatus is configured to decide that the segment of the input audio signal comprises voiced speech as a result of determining that the peak height exceeds a first threshold and the peak width is less than a second threshold.

11. The apparatus of claim 9, wherein the apparatus is configured to decide that the segment of the input audio signal does not comprise voiced speech as a result of determining that the peak height exceeds a first threshold and the peak width exceeds a second threshold.

12. The apparatus of claim 9, wherein the apparatus is configured to determine the peak width by performing a process that includes:

calculating number of bins upwards from the middle of the peak before the ACF curve falls below a fall-off threshold;

calculating number of bins downwards from the middle of the peak before the ACF curve falls below said fall-off threshold; and

adding the numbers of calculated bins to indicate the peak width.

13. The apparatus of claim 9, wherein the apparatus is comprised in:

a server, a client, a network node, a cloud entity or a user equipment.

14. The apparatus of claim 9, wherein the apparatus is comprised in a voice activity detector.

15. A detector apparatus for audio signal processing, the detector apparatus being configured to:

calculate a correlation function of a portion of an input audio signal;

detect a highest peak of said correlation function;

determine a peak width of said peak;

determine a peak height of said peak;

compare the peak width with a first threshold;

compare the peak height with a second threshold; and

decide based on the peak width and the peak height whether a segment of the input audio signal comprises voiced speech.

16. The detector apparatus of claim 15, wherein the detector apparatus is configured to decide that the segment of the input audio signal comprises voiced speech as a result of determining that the peak height exceeds a first threshold and the peak width is less than a second threshold.

17. The detector apparatus of claim 15, wherein the detector apparatus is configured to decide that the segment of the input audio signal does not comprise voiced speech as a result of determining that the peak height exceeds a first threshold and the peak width exceeds a second threshold.

18. The detector apparatus of claim 15, wherein the detector apparatus is configured to determine the peak width by performing a process that includes:

calculating number of bins upwards from the middle of the peak before the ACF curve falls below a fall-off threshold;

calculating number of bins downwards from the middle of the peak before the ACF curve falls below said fall-off threshold; and

adding the numbers of calculated bins to indicate the peak width.