Frequency segmentation to obtain bands for efficient coding of digital media

• US 20070016412A1
• Filed: 07/15/2005
• Published: 01/18/2007
• Est. Priority Date: 07/15/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Chen, Wei-Ge, Mehrotra, Sanjeev
• Granted Patent: US 7,630,882 B2
• Time in Patent Office: Days
• US Class Current: 704/229
• CPC Class Codes: G10L 19/0208 Subband vocoders
• 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

Abstract

Frequency segmentation is important to the quality of encoding spectral data. Segmentation involves breaking the spectral data into units called sub-bands or vectors. Homogeneous segmentation may be suboptimal. Various features are described for providing spectral data intensity dependent segmentation. Finer segmentation is provided for regions of greater spectral variance and coarser segmentation is provided for more homogeneous regions. Sub-bands which have similar characteristics may be merged with very little effect on quality, whereas sub-bands with highly variable data may be better represented if a sub-band is split. Various methods are described for measuring tonality, energy, or shape of a sub-band. These various measurements are discussed in light of making decisions of when to split or merge sub-bands to provide variable frequency segmentation.

Claims

1. An audio encoding method, comprising:

transforming an audio signal into spectral data;

coding a baseband portion of the spectral data;

in an extended band portion of the spectral data, determining characteristics of spectral data;

coding an altered configuration of sub-bands comprising data indicating individual sub-bands in the extended band altered from an initial configuration.

2. The audio encoding method of claim 1 wherein the spectral data comprises coefficients in a transform domain and the altered configuration comprises difference values for sub-bands altered in size from the initial or default configuration.

3. The audio encoding method of claim 1 wherein the initial configuration is a uniform split configuration or a non-uniform split configuration.

4. The audio encoding method of claim 2 wherein a first bit is provided for coding whether a band configuration is default or optimized, and a second bit is provided for coding whether the initial configuration is the uniform split configuration or the non-uniform split configuration.

5. The audio encoding method of claim 1 wherein the altered configuration comprise sub-band multipliers that reflect the relative ratio of a sub-band size to the smallest sub-band size.

6. The audio encoding method of claim 1 wherein the altered configuration comprises sub-band multipliers reflecting splits and merges of sub-bands from the initial configuration.

7. The audio encoding method of claim 1 wherein characteristics of spectral data comprise a measure of at least one of tonality, energy, or shape.

8. The audio encoding method of claim 1 wherein the initial configuration is altered at least in part based on tonality, and the method further comprises:

transforming the audio signal into fast fourier transform blocks;

time averaging adjacent fast fourier transform blocks;

determining a median filtered value by median filtering the time averaged adjacent fast fourier transform blocks;

comparing the time averaged adjacent fast fourier transform blocks to the median filtered value to obtain a tonality number;

determining a corresponding sub-band related to the adjacent fast fourier transform blocks; and

assigning a tonal characteristic to the corresponding sub-band if the tonality number is above a threshold which can be represented by an absolute number, a given percentage of the median filtered value, or a percentage of a local standard deviation of the median filtered value.

9. The audio encoding method of claim 8 wherein the tonal characteristic is at least one of the factors used to determine whether or not to split or merge the corresponding sub-band.

10. The audio encoding method of claim 1 wherein a ratio of energy in adjacent sub-bands is at least partially determinative of whether or not to alter the initial configuration.

11. The audio encoding method of claim 1 wherein sub-band shape differentiation is at least partially determinative of whether or not to split a sub-band.

12. The audio encoding method of claim 1 wherein a decision to split an individual sub-band into two sub-bands is at least partially made when the two split sub-bands have a means-square Euclidean difference that is lower than the individual sub-band by a threshold amount.

13. The audio encoding method of claim 1 wherein coding the altered configuration further comprises coding a minimum ratio sub-band size.

14. An output bit-stream created using the method of claim 1.

15. A decoder decoding an output of claim 1.

16. An audio decoding method, comprising:

decoding an encoded baseband;

decoding an encoded extended band comprising,

receiving data comprising a minimum ratio sub-band size and an altered configuration,

determining a smallest sub-band size in the altered configuration by dividing the smallest sub-band size in the default configuration by the minimum ratio sub-band size, and

determining the actual sub-band multiplier by adding an expected sub-band multiplier to a coded difference value.

17. The audio decoding method of claim 16 wherein the initial configuration is a non-uniform split configuration.

18. The audio decoding method of claim 16 wherein for a second sub-band, data received indicates no alteration from the initial configuration and the second sub-band is decoded according to the initial configuration.

19. An audio encoder comprising:

a transformer for transforming an audio signal into spectral data;

a base coder for coding a baseband portion of the spectral data;

an extended band coder for,

configuring variable sized sub-bands based on characteristics of spectral data in an extended band

coding difference values indicating how individual sub-bands differ in size from an initial configuration,

coding a minimum ratio sub-band size, and

coding sub-bands in the extended band.

20. The audio encoder of claim 19 wherein difference values are determined at least in part by sub-band split or merge from the initial configuration.