Reconstructing an audio signal by spectral component regeneration and noise blending

• US 9,177,564 B2
• Filed: 05/31/2013
• Issued: 11/03/2015
• Est. Priority Date: 03/28/2002
• Status: Active Grant

• Current Assignee: Dolby Laboratories Licensing Corporation (Dolby Laboratories Incorporated)
• Original Assignee: Dolby Laboratories Licensing Corporation (Dolby Laboratories Incorporated)
• Inventors: Truman, Michael Mead, Vinton, Mark Stuart
• Primary Examiner(s): YEN, ERIC L
• Application Number:
  US13/906,994
• Publication Number:
  US 20140161283A1
• Time in Patent Office: 886 Days
• US Class Current: 1/1
• CPC Class Codes: G10L 19/0017 Lossless audio signal coding; Perfect reconstruction of coded audio signal by transmission of coding error G10L19/24 takes precedence
• CPC Class Codes: G10L 19/002 Dynamic bit allocation for perceptual audio coders G10L19/032
• CPC Class Codes: G10L 19/012 Comfort noise or silence coding
• CPC Class Codes: G10L 19/02 using spectral analysis, e.g. transform vocoders or subband vocoders
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: G10L 19/0208 Subband vocoders
• CPC Class Codes: G10L 19/0212 using orthogonal transformation
• CPC Class Codes: G10L 19/028 Noise substitution, i.e. substituting non-tonal spectral components by noisy source comfort noise for discontinuous speech transmission G10L19/012
• CPC Class Codes: G10L 19/03 Spectral prediction for preventing pre-echo; Temporary noise shaping [TNS], e.g. in MPEG2 or MPEG4
• CPC Class Codes: G10L 19/06 Determination or coding of the spectral characteristics, e.g. of the short-term prediction coefficients
• CPC Class Codes: G10L 19/16 Vocoder architecture
• CPC Class Codes: G10L 19/167 Audio streaming, i.e. formatting and decoding of an encoded audio signal representation into a data stream for transmission or storage purposes
• CPC Class Codes: G10L 19/173 Transcoding, i.e. converting between two coded representations avoiding cascaded coding-decoding
• CPC Class Codes: G10L 19/26 Pre-filtering or post-filtering
• CPC Class Codes: G10L 19/265 Pre-filtering, e.g. high frequency emphasis prior to encoding
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence
• CPC Class Codes: G10L 21/038 using band spreading techniques
• CPC Class Codes: G10L 21/0388 Details of processing therefor

Abstract

An audio signal is conveyed more efficiently by transmitting or recording a baseband of the signal with an estimated spectral envelope and a noise-blending parameter derived from a measure of the signal'"'"'s noise-like quality. The signal is reconstructed by translating spectral components of the baseband signal to frequencies outside the baseband, adjusting phase of the regenerated components to maintain phase coherency, adjusting spectral shape according to the estimated spectral envelope, and adding noise according to the noise-blending parameter. Preferably, the transmitted or recorded signal also includes an estimated temporal envelope that is used to adjust the temporal shape of the reconstructed signal.

Claims

1. A method, performed by one or more processors, for generating a time-domain representation of a reconstructed signal that comprises:

receiving a formatted signal comprising;

baseband transform coefficients of an audio signal, wherein the baseband transform coefficients are Modified Discrete Cosine Transform coefficients,

a noise blending parameter, and

information regarding a spectral envelope of the audio signal, wherein the information regarding the spectral envelope of the audio signal is indicative of an estimate of the spectral envelope of the audio signal;

extracting, from the formatted signal;

the baseband transform coefficients,

the noise blending parameter, and

the information regarding the spectral envelope of the audio signal;

generating noise-signal transform coefficients, wherein the noise-signal transform coefficients are weighted in amplitude by a noise blending function that is a function of frequency and the noise blending parameter and that gives greater weight to transform coefficients corresponding to higher frequencies;

generating regenerated-signal transform coefficients that;

correspond to frequencies above the baseband transform coefficients,

are copied from a subset of the baseband transform coefficients, and

are weighted in amplitude by an inverse of the noise blending function;

generating noisy regenerated transform coefficients by combining;

the regenerated-signal transform coefficients, and

the noise-signal transform coefficients,

wherein amplitudes of the noisy regenerated transform coefficients are adjusted in response to the information regarding the spectral envelope of the audio signal;

generating a frequency-domain representation of the reconstructed signal by combining;

the baseband transform coefficients, and

the noisy regenerated transform coefficients;

generating the time-domain representation of the reconstructed signal by applying an inverse Modified Discrete Cosine Transform to the frequency-domain representation of the reconstructed signal; and

generating, from the time-domain representation of the reconstructed signal, an acoustical representation of the reconstructed signal.

2. The method of claim 1, wherein the reconstructed signal comprises segments that vary in length.

3. The method of claim 1, wherein the subset of the baseband transform coefficients excludes baseband transform coefficients for frequencies below 1 kHz.

4. The method of claim 1 that comprises adapting the generating of the regenerated signal by changing which baseband transform coefficients are copied or by changing the frequency amount by which baseband transform coefficients are copied.

5. The method of claim 1, wherein:

the formatted signal also comprises one or more additional noise blending parameters; and

the noise blending function is also a function of the one or more additional noise blending parameters.

6. The method of claim 1, wherein the regenerated signal comprises regenerated-signal transform coefficients copied from the baseband transform coefficients in a circular manner.

7. An apparatus for generating a time-domain representation of a reconstructed signal, wherein the apparatus comprises one or more processors configured to:

receive a formatted signal comprising;

baseband transform coefficients of an audio signal, wherein the baseband transform coefficients are Modified Discrete Cosine Transform coefficients,

a noise blending parameter, and

information regarding a spectral envelope of the audio signal, wherein the information regarding the spectral envelope of the audio signal is indicative of an estimate of the spectral envelope of the audio signal;

extract, from the formatted signal;

the baseband transform coefficients,

the noise blending parameter, and

the information regarding the spectral envelope of the audio signal;

generate noise-signal transform coefficients, wherein the noise-signal transform coefficients are weighted in amplitude by a noise blending function that is a function of frequency and the noise blending parameter and that gives greater weight to transform coefficients corresponding to higher frequencies;

generate regenerated-signal transform coefficients that;

correspond to frequencies above the baseband transform coefficients,

are copied from a subset of the baseband transform coefficients, and

are weighted in amplitude by an inverse of the noise blending function;

generate noisy regenerated transform coefficients by combining;

the regenerated-signal transform coefficients, and

the noise-signal transform coefficients,

wherein amplitudes of the noisy regenerated transform coefficients are adjusted in response to the information regarding the spectral envelope of the audio signal;

generate a frequency-domain representation of the reconstructed signal by combining;

the baseband transform coefficients, and

the noisy regenerated transform coefficients;

generate the time-domain representation of the reconstructed signal by applying an inverse Modified Discrete Cosine Transform to the frequency-domain representation of the reconstructed signal; and

wherein the apparatus is further configured to generate, from the time-domain representation of the reconstructed signal, an acoustical representation of the reconstructed signal.

8. The apparatus of claim 7, wherein the reconstructed signal comprises segments that vary in length.

9. The apparatus of claim 7, wherein the subset of the baseband transform coefficients excludes baseband transform coefficients for frequencies below 1 kHz.

10. The apparatus of claim 7 that comprises means for adapting the generating of the regenerated signal by changing which baseband transform coefficients are copied or by changing the frequency amount by which baseband transform coefficients are copied.

11. The apparatus of claim 7, wherein:

the formatted signal also comprises one or more additional noise blending parameters; and

the noise blending function is also a function of the one or more additional noise blending parameters.

12. The apparatus of claim 7, wherein the regenerated signal comprises regenerated-signal transform coefficients copied from the baseband transform coefficients in a circular manner.

13. A non-transitory medium that is readable by a device and that records a program of instructions executable by the device to perform a method for generating a time-domain representation of a reconstructed signal, wherein the method comprises:

receiving a formatted signal comprising;

baseband transform coefficients of an audio signal, wherein the baseband transform coefficients are Modified Discrete Cosine Transform coefficients,

a noise blending parameter, and

information regarding a spectral envelope of the audio signal, wherein the information regarding the spectral envelope of the audio signal is indicative of an estimate of the spectral envelope of the audio signal;

extracting, from the formatted signal;

the baseband transform coefficients,

the noise blending parameter, and

the information regarding the spectral envelope of the audio signal;

generating noise-signal transform coefficients, wherein the noise-signal transform coefficients are weighted in amplitude by a noise blending function that is a function of frequency and the noise blending parameter and that gives greater weight to transform coefficients corresponding to higher frequencies;

generating regenerated-signal transform coefficients that;

correspond to frequencies above the baseband transform coefficients,

are copied from a subset of the baseband transform coefficients, and

are weighted in amplitude by an inverse of the noise blending function;

generating noisy regenerated transform coefficients by combining;

the regenerated-signal transform coefficients, and

the noise-signal transform coefficients,

wherein amplitudes of the noisy regenerated transform coefficients are adjusted in response to the information regarding the spectral envelope of the audio signal;

generating a frequency-domain representation of the reconstructed signal by combining;

the baseband transform coefficients, and

the noisy regenerated transform coefficients;

generating the time-domain representation of the reconstructed signal by applying an inverse Modified Discrete Cosine Transform to the frequency-domain representation of the reconstructed signal; and

generating, from the time-domain representation of the reconstructed signal, an acoustical representation of the reconstructed signal.

14. The medium of claim 13, wherein the reconstructed signal comprises segments that vary in length.

15. The medium of claim 13, wherein the subset of the baseband transform coefficients excludes baseband transform coefficients for frequencies below 1 kHz.

16. The medium of claim 13 that comprises adapting the generating of the regenerated signal by changing which baseband transform coefficients are copied or by changing the frequency amount by which baseband transform coefficients are copied.

17. The medium of claim 13, wherein:

the formatted signal also comprises one or more additional noise blending parameters; and

the noise blending function is also a function of the one or more additional noise blending parameters.

18. The medium of claim 13, wherein the regenerated signal comprises regenerated-signal transform coefficients copied from the baseband transform coefficients in a circular manner.