Scalable audio coding

• US 7,921,007 B2
• Filed: 07/25/2005
• Issued: 04/05/2011
• Est. Priority Date: 08/17/2004
• Status: Expired due to Fees

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Van de Par, Steven Leonardus Josephus Dimphina Elisabeth, Van Schijndel, Nicolle Hanneke, Kot, Valery Stephanovich
• Primary Examiner(s): YEN, ERIC L
• Application Number:
  US11/573,570
• Publication Number:
  US 20070198274A1
• Time in Patent Office: 2,080 Days
• Field of Search: 704/200.1, 704/201, 704/220, 704/223, 704226-229
• US Class Current: 704/200.1
• 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 19/24 Variable rate codecs, e.g. for generating different qualities using a scalable representation such as hierarchical encoding or layered encoding
• CPC Class Codes: G10L 21/038 using band spreading techniques

Abstract

The invention relates to an audio encoder and decoder and methods for audio encoding and decoding. In a preferred encoder embodiment an audio signal is encoded by deterministic encoder means to form a first encoded signal part. A spectrum of the audio signal is determined and represented by an excitation pattern, i.e. spectral values corresponding to human auditory filters, as a second encoded signal part. A masking curve is also extracted based on the excitation pattern, thus improving encoding efficiency in terms of bit rate. In a preferred decoder the first encoded signal part is decoded by deterministic decoder means. A noise generator uses the decoded first signal part together with the second signal part, i.e. the excitation pattern for the original audio signal, to generate a noise signal. The noise signal is then added to the first decoded signal part to form an output audio signal. At the decoder side the masking curve is also extracted based on the second encoded signal part, i.e. the excitation pattern. The noise signal is generated so that the output audio signal exhibits an excitation pattern nearly identical to the original audio signal. Thus, a perceived high quality audio is obtained while the encoded signal is scalable since a possible deviation between encoding and decoding of the first signal part is compensated by the noise generator at the decoder side. In preferred embodiments the coding means comprises a sinusoidal coder.

Claims

1. An audio encoder for encoding an audio signal (IN), the audio encoder comprising:

encoder means (ENC) for encoding the audio signal (IN) into a first encoded signal part; and

computation means for computing a representation of an excitation pattern of the audio signal and providing the representation of the excitation pattern as a second encoded signal part, wherein the representation of the excitation pattern comprises a representation of human auditory nerve response modeled by a filter bank of parallel auditory filters, the filters in the filter bank having values which relate to a signal level of a frequency band of a corresponding auditory filter, the excitation pattern of the audio signal thereby being a parametric spectral description of the audio signal, the computation means further for computing a representation of a masking curve based on quantized excitation parameters of the representation of the excitation pattern, and providing the representation of the masking curve to the encoder means so as to optimize encoding efficiency of the encoder means, wherein the encoder means encodes signal components of the audio signal relative to the masking curve, further wherein the second encoded signal part, included within an output bit stream of the audio encoder, along with the first signal part, provides a scalable encoded audio signal of the audio encoder.

2. The audio encoder according to claim 1, wherein the audio encoder means comprises a deterministic signal type of encoder selected from the group consisting of:

parametric encoders, transform encoders, waveform encoders, Regular Pulse Excitation encoders, and Codebook Excited Linear Predictive encoders.

3. The audio encoder according to claim 1, further comprising:

means for generating a quantized version of the representation of the excitation pattern prior to providing it the representation of the excitation pattern as the second encoded signal part.

4. The audio encoder according to claim 1, further comprising:

means adapted to code the second encoded signal part according to a coding scheme selected from the group consisting of;

intra-frame differential coding and across segment differential coding.

15. Device comprising an audio encoder according to claim 1.

5. An audio decoder for regenerating an audio signal from an encoded audio signal based on an original audio signal, the encoded audio signal including a first encoded audio signal part and a second encoded audio signal part, the audio decoder comprising:

means for generating, from the second encoded audio signal part, a representation of an excitation pattern of the original audio signal, wherein the representation of the excitation pattern comprises a representation of human auditory nerve response modeled by a filter bank of parallel auditory filters, the filters in the filter bank having values which relate to a signal level of a frequency band of a corresponding auditory filter, the excitation pattern of the audio signal thereby being a parametric spectral description of the original audio signal;

decoder means for generating a first decoded signal part from (i) the first encoded signal part and (ii) a masking curve based on quantized excitation parameters of the representation of the excitation pattern; and

signal generator means for generating a second decoded signal part, based on a scalable noise model, in response to the representation of the excitation pattern and the first decoded signal part, so that a sum of the first and second decoded signal parts exhibits an excitation pattern that is substantially equal to the excitation pattern of the original audio signal, for creating a resulting regenerated audio signal with perceivable spectral properties similar to the original audio signal.

6. The audio decoder according to claim 5, further comprising:

summing means for generating a representation of the audio signal as a sum of the first and second decoded signal parts.

7. The audio decoder according to claim 5, wherein the signal generator means comprises means for generating the second decoded signal part based on the representation of the excitation pattern of the original audio signal by using an iterative method.

8. The audio decoder according to claim 5, wherein the signal generator means performs a subtraction of a representation of an excitation pattern of the first decoded signal part from the excitation pattern of the original audio signal.

9. The audio decoder according to claim 5, wherein the signal generator means comprises a noise generator.

10. The audio decoder according to claim 5, wherein the signal generator means comprises spectral band replication means.

11. The audio decoder according to claim 5, wherein the decoder means comprises a deterministic signal type of decoder selected from the group consisting of:

parametric decoders, transform decoders, waveform decoder, Regular Pulse Excitation encoders, and Codebook Excited Linear Predictive encoders.

12. The audio decoder according to claim 5, further comprising means for computing a representation of the masking curve corresponding to the representation of the excitation pattern of the original audio signal and providing the representation of the masking curve to the decoder means.

16. Device comprising an audio decoder according to claim 5.

13. A method of encoding an audio signal comprising the steps of:

computing, in an excitation pattern computation means, a representation of an excitation pattern of the audio signal, wherein the representation of the excitation pattern comprises a representation of human auditory nerve response modeled by a filter bank of parallel auditory filters, having values each of which relate to a signal level of a frequency band of a corresponding auditory filter, providing a parametric spectral description of the audio signal;

computing, in a masking curve computation unit, a representation of a masking curve based on quantized excitation parameters of the representation of the excitation pattern;

encoding, using encoding means, the audio signal according to an encoding scheme into a first encoded signal part by utilizing the masking curve so as to optimize an encoding efficiency of the encoding, wherein the encoding encodes signal components of the audio signal relative to the masking curve; and

providing, using the excitation pattern computation means, a second encoded signal part comprising the representation of the excitation pattern of the audio signal, wherein the second encoded signal part, for being included within an output bit stream, along with the first signal part, provides a scalable encoded audio signal.

17. A non-transitory computer-readable storage medium embodied with computer program code for being loaded into a memory and executed by a signal processor for encoding an audio signal according to the method of claim 13.

14. A method of regenerating an audio signal from an encoded audio signal based on an original audio signal, the encoded audio signal including a first encoded signal part and a second encoded signal part, the method comprising the steps of:

generating, using a noise modeler, from the second encoded signal part, a representation of an excitation pattern of the original audio signal, wherein the representation of the excitation pattern comprises a representation of human auditory nerve response modeled by a filter bank of parallel auditory filters, having values each of which relate to a signal level of a frequency band of a corresponding auditory filter, providing a parametric spectral description of the original audio signal;

generating, using a masking curve computation unit, from the representation of the excitation pattern, a representation of a masking curve, the masking curve based on quantized excitation parameters of the representation of the excitation pattern;

decoding, using decoding means, a first encoded signal part, according to a decoding scheme, into a first decoded signal part, wherein the decoding includes using the masking curve to decode the first encoded signal part; and

generating, using the noise modeler, a second decoded signal part, based on a scalable noise model, in response to the representation of the excitation pattern and the first decoded signal part, so that a sum of the first and second decoded signal parts exhibits an excitation pattern that is substantially equal to the excitation pattern of the original audio signal, for creating a resulting regenerated audio signal with perceivable spectral properties similar to the original audio signal.

18. A non-transitory computer-readable storage medium embodied with computer program code for being loaded into a memory and executed by a signal processor for decoding by regenerating an audio signal from an encoded audio signal according to the method of claim 14.