Adaptive High-Pass Post-Filter

• US 20150051905A1
• Filed: 08/13/2014
• Published: 02/19/2015
• Est. Priority Date: 08/15/2013
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Inventors: Gao, Yang
• Granted Patent: US 9,418,671 B2
• Time in Patent Office: Days
• US Class Current: 704/207
• CPC Class Codes: G10L 19/125 Pitch excitation, e.g. pitch synchronous innovation CELP [PSI-CELP]
• CPC Class Codes: G10L 19/26 Pre-filtering or post-filtering
• CPC Class Codes: G10L 2019/0011 Long term prediction filters, i.e. pitch estimation

Abstract

In accordance with an embodiment of the present invention, a method of speech processing included receiving a coded audio signal having coding noise. The method further includes generating a decoded audio signal from the coded audio signal, and determining a pitch corresponding to the fundamental frequency of the audio signal. The method also includes determining the minimum allowable pitch and determining if the pitch of the audio signal is less than the minimum allowable pitch. If the pitch of the audio signal is less than the minimum allowable pitch, applying an adaptive high pass filter on the decoded audio signal to lower the coding noise at frequencies below the fundamental frequency.

Claims

1. A method of speech processing using a code excitation linear prediction (CELP) algorithm, the method comprising:

receiving a coded audio signal comprising coding noise;

generating a decoded audio signal from the coded audio signal;

determining a pitch corresponding to a fundamental frequency of the audio signal;

determining a minimum allowable pitch for the CELP algorithm;

determining whether the pitch of the audio signal is less than the minimum allowable pitch; and

when the pitch of the audio signal is less than the minimum allowable pitch, applying an adaptive high pass filter on the decoded audio signal to lower the coding noise at frequencies below the fundamental frequency.

2. The method of claim 1, wherein the adaptive high pass filter is included in a code-excited linear prediction (CELP) decoder.

3. The method of claim 1, when the pitch of the decoded audio signal is greater than the maximum allowable fundamental pitch, not applying the adaptive high pass filter.

4. The method of claim 1, further comprising:

determining whether the audio signal is a voiced speech signal; and

not applying the adaptive high pass filter when the decoded audio signal is determined to be not a voiced speech signal.

5. The method of claim 1, further comprising:

determining whether the audio signal was coded using a CELP encoder; and

not applying the adaptive high pass filter on the decoded audio signal when the decoded audio signal was not coded using a CELP encoder.

6. The method of claim 1, wherein a cut-off frequency of the adaptive high pass filter is less than the fundamental frequency.

7. The method of claim 6, wherein the adaptive high pass filter is a second order high-pass filter.

8. The method of claim 7, wherein the adaptive high pass filter is given by the

9. The method of claim 1, wherein a first subframe of a frame of the coded audio signal is coded in a full range from a minimum pitch limit to a maximum pitch limit, and wherein the minimum allowable pitch is the minimum pitch limit of the CELP algorithm.

10. A method of speech processing using a code excitation linear prediction (CELP) algorithm, the method comprising:

receiving a voiced wideband spectrum comprising coding noise;

determining a pitch corresponding to a fundamental frequency of the voiced wideband spectrum;

determining a minimum allowable pitch for the CELP algorithm;

determining whether the pitch of the voiced wideband spectrum is less than the minimum allowable pitch; and

when the pitch of the voiced wideband spectrum is less than the minimum allowable pitch, applying an adaptive high pass filter having a cut-off frequency less than the fundamental frequency on the voiced wideband spectrum to lower the coding noise at frequencies below the fundamental frequency.

11. The method of claim 10, wherein the voiced wideband spectrum is a synthesized speech output of a code-excited linear prediction (CELP) decoder.

12. The method of claim 10, wherein the adaptive high pass filter is configured to not modify the voiced wideband spectrum when the pitch of the voiced wideband spectrum is greater than the minimum allowable pitch.

13. The method of claim 10, further comprising:

determining whether the voiced wideband spectrum was coded using a CELP encoder; and

wherein the adaptive high pass filter is configured to not modify the voiced wideband spectrum when the voiced wideband spectrum was not coded using a CELP encoder.

14. The method of claim 10, wherein the cut-off frequency of the adaptive high pass filter is less than the fundamental frequency.

15. The method of claim 14, wherein the adaptive high pass filter is a second order high-pass filter.

16. The method of claim 15, wherein the adaptive high pass filter is given by the

17. A code excitation linear predictive (CELP) decoder comprising:

an excitation codebook for outputting a first excitation signal of a speech signal;

a first gain stage for amplifying the first excitation signal from the excitation codebook;

an adaptive codebook for outputting a second excitation signal of the speech signal;

a second gain stage for amplifying the second excitation signal from the adaptive codebook;

a adder for adding the amplified first excitation code vector with the amplified second excitation code vector;

a short term prediction filter configured to filter the output of the adder and output a synthesized speech signal; and

an adaptive high pass filter coupled to the output of the short term prediction filter, the adaptive high filter comprising an adjustable cut-off frequency to dynamically filter out coding noise below the fundamental frequency in the synthesized speech signal.

18. The CELP decoder of claim 17, wherein the adaptive high pass filter is configured to not modify the synthesized speech signal when the fundamental frequency of the synthesized speech signal is less than the maximum allowable fundamental frequency.

19. The CELP decoder of claim 17, wherein the adaptive high pass filter is configured to not modify the synthesized speech signal when the speech signal was not coded using a CELP encoder.

20. The CELP decoder of claim 17, wherein the adaptive high pass filter is given by the