Time-warping frames of wideband vocoder

• US 8,239,190 B2
• Filed: 08/22/2006
• Issued: 08/07/2012
• Est. Priority Date: 08/22/2006
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Kapoor, Rohit, Spindola, Serafin Diaz
• Primary Examiner(s): Desir, Pierre-Louis
• Assistant Examiner(s): KOVACEK, DAVID M
• Application Number:
  US11/508,396
• Publication Number:
  US 20080052065A1
• Time in Patent Office: 2,177 Days
• Field of Search: 704200-201, 704/203, 704206-211, 704214-223, 704226-229, 704/231, 704236-245, 704/258, 704262-269, 704E15001-E15002, 704E19001-E19008, 704/E19.01, 704E19028-E19029, 704E19035-E19038, 704/E21.001, 704/E21.011, 704E21017-E21018
• US Class Current: 704/203
• CPC Class Codes: G10L 19/087 using mixed excitation models, e.g. MELP, MBE, split band LPC or HVXC
• CPC Class Codes: G10L 19/18 Vocoders using multiple modes
• CPC Class Codes: G10L 21/01 Correction of time axis

Abstract

A method of communicating speech comprising time-warping a residual low band speech signal to an expanded or compressed version of the residual low band speech signal, time-warping a high band speech signal to an expanded or compressed version of the high band speech signal, and merging the time-warped low band and high band speech signals to give an entire time-warped speech signal. In the low band, the residual low band speech signal is synthesized after time-warping of the residual low band signal while in the high band, an unwarped high band signal is synthesized before time-warping of the high band speech signal. The method may further comprise classifying speech segments and encoding the speech segments. The encoding of the speech segments may be one of code-excited linear prediction, noise-excited linear prediction or ⅛ frame (silence) coding.

Claims

1. A method of communicating speech, comprising:

time-warping a residual low band speech signal to an expanded or compressed version of the residual low band speech signal;

time-warping a high band speech signal to an expanded or compressed version of the high band speech signal, wherein the time-warping of the high band speech signal comprises;

determining a plurality of pitch periods from the residual low band speech signal;

overlap/adding one or more pitch periods of the high band speech signal if the high band speech signal is compressed, using the pitch periods from the residual low band speech signal; and

overlap/adding or repeating one or more pitch periods of the high band speech signal if the high band speech signal is expanded, using the pitch periods from the residual low band speech signal; and

merging a synthesized version of the time-warped residual low band and the time-warped high band speech signals to give an entire time-warped speech signal.

2. The method of claim 1, further comprising synthesizing the time-warped residual low band speech signal.

3. The method of claim 2, further comprising synthesizing the high band speech signal before time-warping it.

4. The method of claim 3, further comprising:

classifying speech segments; and

encoding the speech segments.

5. The method of claim 4, wherein encoding the speech segments comprises using code-excited linear prediction, noise-excited linear prediction or ⅛

frame coding.

6. The method of claim 4, wherein the encoding is code-excited linear prediction encoding.

7. The method of claim 6, wherein the time-warping of the residual low band speech signal comprises:

estimating at least one pitch period; and

adding or subtracting at least one of the pitch periods after receiving the residual low band speech signal.

8. The method of claim 6, wherein the time-warping of the residual low band speech signal comprises:

estimating pitch delay;

dividing a speech frame into pitch periods, wherein boundaries of the pitch periods are determined using the pitch delay at various points in the speech frame;

overlap/adding the pitch periods if the residual low band speech signal is compressed; and

overlap/adding or repeating one or more pitch periods if the residual low band speech signal is expanded.

9. The method of claim 8, wherein the estimating of the pitch delay comprises interpolating between a pitch delay of an end of a last frame and an end of a current frame.

10. The method of claim 8, wherein the overlap/adding or repeating one or more of the pitch periods comprises merging the speech segments.

11. The method of claim 10, further comprising selecting similar speech segments, wherein the similar speech segments are merged.

12. The method of claim 10, further comprising correlating the speech segments, whereby similar speech segments are selected.

13. The method of claim 8, wherein the overlap/adding or repeating one or more of the pitch periods if the residual low band speech signal is expanded comprises adding an additional pitch period created from a first pitch segment and a second pitch period segment.

14. The method of claim 13, wherein the adding of an additional pitch period created from a first pitch segment and a second pitch period segment comprises adding the first and second pitch segments such that the first pitch period segment'"'"'s contribution increases and the second pitch period segment'"'"'s contribution decreases.

15. The method of claim 1, wherein the low band represents the band up to and including 4 kHz.

16. The method of claim 1, wherein the high band represents the band from about 3.5 kHz to about 7 kHz.

36. A non-transitory computer-readable medium having machine-readable instructions performing the method according to claim 1.

17. A vocoder having at least one input and at least one output, comprising:

an encoder comprising a filter having at least one input operably connected to the input of the vocoder and at least one output; and

a decoder comprising;

a synthesizer having at least one input operably connected to the at least one output of the encoder and at least one output operably connected to the at least one output of the vocoder; and

a memory, wherein the decoder is adapted to execute software instructions stored in the memory comprising;

time-warping a residual low band speech signal to an expanded or compressed version of the residual low band speech signal;

time-warping a high band speech signal to an expanded or compressed version of the high band speech signal, wherein the time-warping software instruction of the high band speech signal comprises;

determining a plurality of pitch periods from the residual low band speech signal,

overlap/adding one or more pitch periods of the high band speech signal if the high band speech signal is compressed, using the pitch periods from the residual low band speech signal; and

overlap/adding or repeating one or more pitch periods of the high band speech signal if the high band speech signal is expanded, using the pitch periods from the residual low band speech signal; and

merging a synthesized version the time-warped residual low band and the time-warped high band speech signals to give an entire time-warped speech signal.

18. The vocoder of claim 17, wherein the synthesizer comprises means for synthesizing the time-warped residual low band speech signal.

19. The vocoder of claim 18, wherein the synthesizer further comprises means for synthesizing the high band speech signal before time-warping it.

20. The vocoder of claim 19, wherein the encoder comprises a memory and the encoder is adapted to execute software instructions stored in the memory comprising encoding speech segments using code-excited linear prediction encoding.

21. The vocoder of claim 20, wherein the time-warping software instruction of the high band speech signal comprises:

overlap/adding the same number of samples as were compressed in the lower band if the high band speech signal is compressed; and

overlap/adding the same number of samples as were expanded in the lower band if the high band speech signal is expanded.

22. The vocoder of claim 20, wherein the time-warping software instruction of the residual low band speech signal comprises:

estimating at least one pitch period; and

adding or subtracting the at least one pitch period after receiving the residual low band speech signal.

23. The vocoder of claim 20, wherein the time-warping software instruction of the residual low band speech signal comprises:

estimating pitch delay;

dividing a speech frame into pitch periods, wherein boundaries of the pitch periods are determined using the pitch delay at various points in the speech frame;

overlap/adding the pitch periods if the residual speech signal is compressed; and

overlap/adding or repeating one or more pitch periods if the residual speech signal is expanded.

24. The vocoder of claim 23, wherein the overlap/adding instruction of the pitch periods if the residual low band speech signal is compressed comprises:

segmenting an input sample sequence into blocks of samples;

removing segments of the residual signal at regular time intervals;

merging the removed segments; and

replacing the removed segments with a merged segment.

25. The vocoder of claim 24, wherein the merging instruction of the removed segments comprises increasing a first pitch period segment'"'"'s contribution and decreasing a second pitch period segment'"'"'s contribution.

26. The vocoder of claim 23, wherein the estimating instruction of the pitch delay comprises interpolating between a pitch delay of an end of a last frame and an end of a current frame.

27. The vocoder of claim 23, wherein the overlap/adding or repeating one or more of the pitch periods instruction comprises merging the speech segments.

28. The vocoder of claim 27, further comprising selecting similar speech segments, wherein the similar speech segments are merged.

29. The vocoder of claim 27, wherein the time-warping instruction of the residual low band speech signal further comprises correlating the speech segments, whereby similar speech segments are selected.

30. The vocoder of claim 23, wherein the overlap/adding or repeating one or more of the pitch periods instruction if the residual low band speech signal is expanded comprises adding an additional pitch period created from a first pitch period segment and a second pitch period segment.

31. The vocoder of claim 30, wherein the adding instruction of an additional pitch period created from the first and second pitch period segments comprises adding the first and second pitch period segments such that the first pitch period segment'"'"'s contribution increases and the second pitch period segment'"'"'s contribution decreases.

32. The vocoder of claim 17, wherein the encoder comprises a memory and the encoder is adapted to execute software instructions stored in the memory comprising classifying speech segments as ⅛

frame, code-excited linear prediction or noise-excited linear prediction.

33. The vocoder of claim 17, wherein the low band represents the band up to and including 4 kHz.

34. The vocoder of claim 17, wherein the high band represents the band from about 3.5 kHz to about 7 kHz.

35. An apparatus configured to communicate speech, said apparatus comprising:

means for time-warping a residual low band speech signal to an expanded or compressed version of the residual low band speech signal;

means for time-warping a high band speech signal to an expanded or compressed version of the high band speech signal, wherein the time-warping of the high band speech signal comprises;

means for determining a plurality of pitch periods from the residual low band speech signal;

means for overlapping/adding one or more pitch periods of the high band speech signal if the high band speech signal is compressed, using the pitch periods from the residual low band speech signal; and

means for overlapping/adding or repeating one or more pitch periods of the high band speech signal if the high band speech signal is expanded, using the pitch periods from the residual low band speech signal; and

means for merging a synthesized version of the time-warped residual low band and the time-warped high band speech signals to give an entire time-warped speech signal.