Adaptive filter pitch extraction

• US 7,949,520 B2
• Filed: 12/09/2005
• Issued: 05/24/2011
• Est. Priority Date: 10/26/2004
• Status: Active Grant

• Current Assignee: Blackberry Limited
• Original Assignee: QNX Software Systems Company (Blackberry Limited)
• Inventors: Nongpiur, Rajeev, Hetherington, Phillip A.
• Primary Examiner(s): Dorvil; Richemond
• Assistant Examiner(s): Colucci; Michael C
• Application Number:
  US11/298,052
• Publication Number:
  US 20060095256A1
• Time in Patent Office: 1,992 Days
• Field of Search: 381/92, 704/219, 704/208, 704/207, 704/271, 704/258, 704/223, 704/217, 708/300, 705/42, 375/240
• US Class Current: 704/207
• CPC Class Codes: G06Q 20/108 Remote banking, e.g. home banking
• CPC Class Codes: G10L 2021/02085 Periodic noise
• CPC Class Codes: G10L 25/90 Pitch determination of speech signals
• CPC Class Codes: H03H 21/0012 Digital adaptive filters

Abstract

An enhancement system extracts pitch from a processed speech signal. The system estimates the pitch of voiced speech by deriving filter coefficients of an adaptive filter and using the obtained filter coefficients to derive pitch. The pitch estimation may be enhanced by using various techniques to condition the input speech signal, such as spectral modification of the background noise and the speech signal, and/or reduction of the tonal noise from the speech signal.

Claims

1. A system for extracting pitch from a speech signal having voiced and unvoiced segments, the system comprising:

a discrete input;

a delay unit that digitally delays the speech signal received through the discrete input;

an adaptive filter coupled to the delay unit; and

a pitch estimator that analyzes filter coefficients derived from the adaptive filter to estimate the pitch of the speech signal;

where the pitch estimator comprises a processor configured to locate a position of a peak in the filter coefficients derived from the adaptive filter, calculate a number of samples that the speech signal has been delayed, identify a sampling frequency of the speech signal, calculate a sum of the number of samples that the speech signal has been delayed and the position of a peak in the filter coefficients, and divide the sampling frequency by the sum to estimate the pitch of the speech signal.

2. The system of claim 1 further comprising a low-pass filter coupled between the delay unit and the adaptive filter.

3. The system of claim 1 further comprising spectral modification logic coupled between the delay unit and the adaptive filter.

4. The system of claim 1 where the delay unit comprises a programmable delay that delays passing the input signal to the adaptive filter.

5. The system of claim 1 further comprising a programmable filter coupled to the delay unit, where the programmable filter comprises a transfer function functionally related to a transfer function of the adaptive filter.

6. The system of claim 5 where the delay unit comprises a programmable delay that delays passing the input signal to the programmable filter.

7. The system of claim 1 where the system is configured to add a noise to the speech signal before passing the signal through the adaptive filter.

8. The system of claim 7 where the noise comprises a white noise.

19. The system of claim 1 where the filter coefficients define a transfer function of the adaptive filter.

9. A system for extracting pitch from a signal, the system comprising:

a discrete input;

a delay unit that digitally delays the signal received through the discrete input;

an adaptive filter coupled to the delay unit;

a spectral modification logic coupled between the delay unit and the adaptive filter, where the spectral modification logic is configured to modify the spectrum of the signal prior to filtering of the signal by the adaptive filter; and

a pitch estimator that analyzes filter coefficients derived from the adaptive filter to estimate the pitch of the signal;

where the pitch estimator comprises a processor configured to locate a position of a peak in the filter coefficients derived from the adaptive filter, calculate a number of samples that the signal has been delayed, identify a sampling frequency of the signal, calculate a sum of the number of samples that the signal has been delayed and the position of a peak in the filter coefficients, and divide the sampling frequency by the sum to estimate the pitch of the signal.

10. The system of claim 9 further comprising a programmable filter coupled to the delay unit, where the programmable filter comprises a transfer function functionally related to a transfer function of the adaptive filter.

11. The system of claim 10 where the delay unit comprises a programmable delay that delays passing the input signal to the programmable filter.

12. The system of claim 9 where the system is configured to add a noise to the reference input signal before passing the signal through the adaptive filter.

20. The system of claim 9 where the filter coefficients define a transfer function of the adaptive filter.

13. A method for extracting pitch from a speech signal, the method comprising:

delaying the speech signal;

passing the delayed speech signal through an adaptive filter;

obtaining filter coefficients of the adaptive filter; and

analyzing the filter coefficients to estimate pitch in the speech signal;

where the act of analyzing the filter coefficients comprises;

locating, by a processor, a position of a peak in the filter coefficients of the adaptive filter;

calculating a number of samples that the speech signal has been delayed;

identifying a sampling frequency of the speech signal;

calculating a sum of the number of samples that the speech signal has been delayed and the position of a peak in the filter coefficients; and

dividing the sampling frequency by the sum to estimate the pitch of the speech signal by the processor.

14. The method of claim 13 further comprising enhancing the periodic components of the speech signal.

15. The method of claim 13 further comprising reducing the tonal noise in the speech signal.

16. The method of claim 13 further comprising passing the speech signal through a low-pass filter before passing the signal through the adaptive filter.

17. The method of claim 13 further comprising spectrally flattening the speech signal before passing the signal through the adaptive filter.

18. The method of claim 13 further comprising adding noise to the speech signal to enhance an estimation of the filter coefficients.

21. The method of claim 13 where the filter coefficients define a transfer function of the adaptive filter.

22. The method of claim 13 where the act of delaying comprises delaying the speech signal with delay logic, where the act of passing comprises passing the delayed speech signal through the adaptive filter coupled to the delay logic, where the act of analyzing the filter coefficients comprises providing the filter coefficients to a processor that analyzes the filter coefficients to generate an estimate of the pitch in the speech signal.

30. The method of claim 13 where the adaptive filter comprises circuitry, program instructions stored in memory, or both.

23. A system for extracting pitch from a speech signal having voiced and unvoiced segments, the system comprising:

a discrete input;

a delay unit that digitally delays the speech signal received through the discrete input; and

an adaptive filter coupled to the delay unit, where filter coefficients derived from the adaptive filter are used to estimate the pitch of the speech signal;

where the system is configured to calculate a leaky average vector of the filter coefficients, and then subtract the leaky average from a total estimate of the adaptive filter coefficients.

24. The system of claim 23 further comprising a processor configured to analyze the filter coefficients and estimate the pitch of the speech signal based on a position of a peak in the filter coefficients.

31. The system of claim 23 further comprising a pitch estimator configured to analyze the filter coefficients to estimate the pitch of the signal.

32. The system of claim 23 further comprising a pitch estimator that comprises a processor configured to locate a position of a peak in the filter coefficients derived from the adaptive filter, calculate a number of samples that the speech signal has been delayed, identify a sampling frequency of the speech signal, calculate a sum of the number of samples that the speech signal has been delayed and the position of a peak in the filter coefficients, and divide the sampling frequency by the sum to estimate the pitch of the speech signal.

25. A system for extracting pitch from a signal, the system comprising:

a discrete input;

a delay unit that digitally delays the signal received through the discrete input;

an adaptive filter coupled to the delay unit, where filter coefficients derived from the adaptive filter are used to estimate the pitch of the signal; and

a spectral modification logic coupled between the delay unit and the adaptive filter, where the spectral modification logic is configured to modify the spectrum of the signal prior to filtering of the signal by the adaptive filter;

where the system is configured to calculate a leaky average vector of the filter coefficients and subtract the leaky average from an estimate of the adaptive filter coefficients.

26. The system of claim 25 further comprising a processor configured to analyze the filter coefficients and estimate the pitch of the signal based on a position of a peak in the filter coefficients.

27. A method for extracting pitch from a speech signal, the method comprising:

delaying the speech signal;

passing the delayed speech signal through an adaptive filter;

obtaining filter coefficients of the adaptive filter;

utilizing the filter coefficients to estimate pitch in the speech signal;

calculating, by a processor, a leaky average of the filter coefficients that corresponds to a tonal noise; and

subtracting the leaky average vector from the total estimate of the filter coefficients.

28. The method of claim 27 where the act of utilizing the filter coefficients comprises:

locating a position of a peak in the filter coefficients; and

estimating pitch in the speech signal based on the position of the peak.

29. The method of claim 27 where the act of delaying comprises delaying the speech signal with delay logic, and where the act of utilizing the filter coefficients comprises providing the filter coefficients to a processor that analyzes the filter coefficients to generate an estimate of the pitch in the speech signal.

33. The method of claim 27 where the adaptive filter comprises circuitry, program instructions stored in memory, or both.