Packet loss concealment for bandwidth extension of speech signals

• US 9,280,978 B2
• Filed: 03/27/2013
• Issued: 03/08/2016
• Est. Priority Date: 03/27/2012
• Status: Active Grant

• Current Assignee: Gwangju Institute of Science and Technology
• Original Assignee: Gwangju Institute of Science and Technology
• Inventors: Kim, Hong-Kook, Park, Nam-In
• Primary Examiner(s): Lerner, Martin
• Application Number:
  US13/851,245
• Publication Number:
  US 20130262122A1
• Time in Patent Office: 1,077 Days
• Field of Search: 704/200.1, 704/203, 704/205, 704/208, 704/214, 704/500, 704/501
• US Class Current: 1/1
• CPC Class Codes: G10L 19/005 Correction of errors induced by the transmission channel, if related to the coding algorithm
• CPC Class Codes: G10L 19/02 using spectral analysis, e.g. transform vocoders or subband vocoders
• CPC Class Codes: G10L 19/0212 using orthogonal transformation
• CPC Class Codes: G10L 21/0388 Details of processing therefor

Abstract

Disclosed is a speech receiving apparatus. A low-band PLC module and a synthesis filter reconstructs a low-band speech signal of a lost frame from a previous good frame. A high-band PLC module reconstructs a high-band speech signal of the lost frame from the previous good frame. A transforming part transforms the low-band speech signal into a frequency range. A bandwidth extending part generates at least an extended MDCT coefficient as information for the high-band speech signal from the low-band speech signal transformed by the transforming part. A smoothing part smoothes the extended MDCT coefficient. An inverse transforming part inversely transforms the extended MDCT coefficient smoothed by the smoothing part to a time domain. A synthesizing part synthesizes the low-band speech signal, and the high-band speech signal which is inverse-transformed by the inverse transforming part and reconstructed, to output a wideband speech signal.

Claims

1. A speech receiving apparatus comprising:

a low-band packet loss concealment (PLC) module and a synthesis filter reconstructing a low-band speech signal of a lost frame from a previous good frame;

a high-band PLC module reconstructing a high-band speech signal of the lost frame from the previous good frame;

a transforming part transforming the low-band speech signal to a frequency domain;

a bandwidth extending part generating at least an extended modified discrete cosine transform (MDCT) coefficient as information for the high-band speech signal from the low-band speech signal transformed by the transforming part;

a smoothing part smoothing the extended MDCT coefficient;

an inverse transforming part inversely transforming the extended MDCT coefficient smoothed by the smoothing part to a time domain; and

a synthesizing part synthesizing the low-band speech signal, and the high-band speech signal that is inverse-transformed by the inverse transforming part and reconstructed, to output a wideband speech signal;

wherein the bandwidth extending part performs spectral folding of low-band MDCT coefficients to generate at least a part of the extended MDCT coefficients.

2. The speech receiving apparatus of claim 1, wherein the bandwidth extending part comprises at least two processing parts generating the extended MDCT coefficient by a different process according to the frequency range.

3. The speech receiving apparatus of claim 1, wherein the bandwidth extending part comprises a spectral folding part and a spectral smoothing part, generating at least a part of the extended MDCT coefficients by folding and smoothing the MDCT coefficients of the low-band speech signal.

4. The speech receiving apparatus of claim 1, wherein the bandwidth extending part comprises a voiced/unvoiced speech determining part utilizing the MDCT coefficients of the low-band speech signal by different processes according to a voiced or unvoiced speech.

5. The speech receiving apparatus of claim 4, wherein the bandwidth extending part comprises a voiced speech processing part performing a harmonic spectral folding when an input speech is determined to be the voiced speech by the voiced/unvoiced speech determining part.

6. The speech receiving apparatus of claim 4, wherein the bandwidth extending part comprises an unvoiced speech processing part performing a spectral folding of a high autocorrelation section from the low band when an input speech is determined to be the unvoiced speech by the voiced/unvoiced speech determining part.

7. The speech receiving apparatus of claim 4, wherein the voiced/unvoiced speech determining part determines the voiced or unvoiced speech according to a tilt of a spectral tilt parameter.

8. The speech receiving apparatus of claim 1, wherein, in the bandwidth extending part,

the extended MDCT coefficient for a second frequency range is generated by folding the MDCT coefficient of the low-band speech signal,

the extended MDCT coefficient for a third frequency range higher than the second frequency range is generated by folding and smoothing the MDCT coefficient of the low-band speech signal,

the extended MDCT coefficient for a first frequency range lower than the second frequency range is generated by differently processing the MDCT coefficient of the low-band speech signal according to whether an input speech is a voiced or unvoiced speech.

9. The speech receiving apparatus of claim 8, wherein the first frequency range is 4-4.6 kHz, the second frequency range is 4.6-5.5 kHz, and the third frequency range is 5.5-7 kHz.

10. The speech receiving apparatus of claim 1, wherein the bandwidth extending part comprises a shaping part shaping the extended MDCT coefficient that is generated by a different process according to the frequency range and then synthesized.

11. A speech receiving method comprising:

reconstructing a low-band speech signal of a lost frame from a previous good frame;

transforming the reconstructed low-band speech signal to a frequency domain to provide a low-band modified discrete cosine transform (MDCT) coefficient;

processing the low-band MDCT coefficient by different methods according to the frequency ranges of the high band, which are classified into at least two cases, to provide an extended MDCT coefficient of a high-band speech signal;

inversely transforming the extended MDCT coefficient to a time domain to reconstruct the high-band speech signal; and

synthesizing the reconstructed high-band speech signal and the low-band speech signal;

wherein a second frequency range that is a part of the extended MDCT coefficients is obtained by folding the low-band MDCT coefficient.

12. The speech receiving method of claim 11, prior to the reconstructing of the high-band speech signal, further comprising smoothing the high-band extended MDCT coefficient using the high-band MDCT coefficient reconstructed in the previous good frame in order to inhibit the high-band extended MDCT coefficients from being abruptly changed.

13. The speech receiving method of claim 11, wherein a third frequency range that is a part of the extended MDCT coefficients and is higher than the second frequency range is obtained by folding and smoothing the low-band MDCT coefficient.

14. The speech receiving method of claim 11, wherein a third frequency range that is a part of the extended MDCT coefficients utilizes the low-band MDCT coefficient by using different methods according to whether an input speech is a voiced or unvoiced speech.

15. The speech receiving method of claim 14, wherein, when the input speech is the voiced speech, the extended MDCT coefficient is obtained by using the low-band MDCT coefficient by a harmonic spectral replication method.

16. The speech receiving method of claim 14, wherein, when the input speech is the unvoiced speech, the extended MDCT coefficient is obtained by using the low-band MDCT coefficient by an autocorrelation spectral replication method.