Method for reduction of aliasing introduced by spectral envelope adjustment in real-valued filterbanks

• US 7,590,543 B2
• Filed: 09/21/2007
• Issued: 09/15/2009
• Est. Priority Date: 09/18/2002
• Status: Expired due to Term

• Current Assignee: Dolby International AB (Dolby Laboratories Incorporated)
• Original Assignee: Coding Technologies Sweden AB
• Inventors: Villemoes, Lars, Kjorling, Kristofer
• Primary Examiner(s): Dorvil; Richemond
• Assistant Examiner(s): Borsetti; Greg A
• Application Number:
  US11/859,521
• Publication Number:
  US 20080010061A1
• Time in Patent Office: 725 Days
• Field of Search: 704/201, 704/500, 704/229, 704/200.1, 370/240, 381/23
• US Class Current: 704/500
• CPC Class Codes: G10L 19/00 Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis in musical instruments G10H
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: G10L 19/025 Detection of transients or attacks for time/frequency resolution switching
• CPC Class Codes: G10L 19/083 the excitation function being an excitation gain G10L25/90 takes precedence
• CPC Class Codes: G10L 19/12 the excitation function being a code excitation, e.g. in code excited linear prediction [CELP] vocoders
• CPC Class Codes: G10L 19/26 Pre-filtering or post-filtering
• CPC Class Codes: G10L 25/12 the extracted parameters being prediction coefficients
• CPC Class Codes: G10L 25/18 the extracted parameters being spectral information of each sub-band
• CPC Class Codes: G10L 25/21 the extracted parameters being power information
• CPC Class Codes: G10L 25/48 specially adapted for particular use
• CPC Class Codes: H03H 17/0266 Filter banks

Abstract

The present invention proposes a new method for improving the performance of a real-valued filterbank based spectral envelope adjuster. By adaptively locking the gain values for adjacent channels dependent on the sign of the channels, as defined in the application, reduced aliasing is achieved. Furthermore, the grouping of the channels during gain-calculation, gives an improved energy estimate of the real valued subband signals in the filterbank.

Claims

1. Apparatus for calculating gain adjustment values for gain adjusting a plurality of audio subband signals generated by filtering an audio signal using an analysis filterbank, the analysis filterbank having subband filters, adjacent subband filters of the filterbank having transition bands overlapping in an overlapping range, comprising:

an examiner for examining an audio subband signal originating from a subband filter and an adjacent audio subband signal originating from an adjacent subband filter, to determine whether the audio subband signal and the adjacent audio subband signal have aliasing generating signal components in the overlapping range to obtain grouped audio subband signals in response to a result of the examination indicating that the audio subband signal and the adjacent audio subband signal have aliasing generating signal components in the overlapping range;

a calculator for calculating a first gain adjustment value and a second gain adjustment value for the grouped adjacent audio subband signals, wherein the calculator is operative

to determine an indication of a reference energy for the grouped audio subband signals and an energy estimate for an energy in the grouped adjacent audio subband signals, and

to calculate the gain adjustment values for the grouped audio subband signals based on the indication of the reference energy and the energy estimate.

2. Apparatus in accordance with claim 1, further comprising a provider for providing a first reference spectral envelope value for the audio subband signal and a second reference spectral envelope value for the adjacent audio subband signal,

in which the calculator is operative to determine, for grouped audio subband signals, a first energy measure indicating a signal energy of the audio subband signal and a second energy measure indicating a signal energy of the adjacent audio subband signal, and

in which the calculator is further operative to calculate the indication of the reference energy as a linear combination of the first reference spectral envelope value and the second reference spectral envelope value and to calculate the energy estimate for an energy in the grouped adjacent audio subband signals as a linear combination of the first energy measure and the second energy measure.

3. Apparatus in accordance with claim 1, which the calculator is operative to calculate the first and the second gain adjustment values such that they differ by less than a predetermined threshold or are equal to each other.

4. Apparatus in accordance with claim 3, in which the predetermined threshold is lower than or equal to 6 dB.

5. Apparatus in accordance with claim 1, further comprising means for providing an unmodified first gain adjustment value for the audio subband signal and an unmodified second gain adjustment value for the adjacent audio subband signal, and

in which the calculator is operative to calculate the first and the second gain adjustment values so that both are larger than or equal to a lower value of the first and the second unmodified gain adjustment values and smaller or equal to a higher value of the first and the second unmodified gain adjustment values.

6. Apparatus in accordance with claim 5, in which the unmodified first gain adjustment value and the unmodified second gain adjustment value are indicative for a spectral envelope of an original signal in a frequency band, wherein the frequency band is to be reconstructed by a spectral band replication.

7. Apparatus in accordance with claim 1, in which the calculator is operative to calculate the first gain adjustment value and the second gain adjustment value based on an average energy of the audio subband signal and the adjacent audio subband signal.

8. Apparatus in accordance with claim 1,

in which the calculator is operative to calculate the first and the second gain adjustment factors according to the following equation:

$g_k<br><br>\left(m\right)=g_{k-1}<br><br>\left(m\right)=\sqrt{\frac{E_k^{\mathrm{ref}}<br><br>\left(m\right)+E_{k-1}^{\mathrm{ref}}<br><br>\left(m\right)}{E_k<br><br>\left(m\right)+E_{k-1}<br><br>\left(m\right)}},$

wherein g<sub>k−

1</sub>(m) is the gain adjustment factor for a first audio subband signal of grouped audio subband signals, wherein g_k(m) is the gain adjustment factor for a second audio subband signal of the grouped audio subband signals, wherein E<sub>k−

1</sub>^ref(m) is the reference energy for the first audio subband signal, wherein E_k^ref(m) is the reference energy for the second audio subband signal, wherein E<sub>k−

1</sub>(m) is the energy estimate of the first audio subband signal, wherein E_k(m) is the energy estimate of the second audio subband signal, and wherein m indicates a point in time.

9. Apparatus in accordance with claim 1, in which an energy estimate for an audio subband signal is calculated by summing of squared audio subband samples of the audio subband signal.

10. Apparatus in accordance with claim 1, in which the examiner is operative to calculate signs of audio subband signals based on coefficients of prediction polynomials for the audio subband signal and the adjacent audio subband signal, and to indicate the result that the audio subband signal and the adjacent audio subband signal have aliasing generating signal components in the overlapping range, when the signs have a predetermined relationship to each other.

11. Apparatus in accordance with claim 10, in which the examiner is operative to apply an auto-correlation method or a co-variance method.

12. Apparatus in accordance with claim 10, in which the prediction polynomial is a low order polynomial having a first order coefficient, wherein the order of the low order polynomial is smaller than 4 and in which the examiner is operative to use the first order coefficient for calculating the signs of the audio subband signals.

13. Apparatus in accordance with claim 10, in which the examiner is operative to calculate the sign for an audio subband signal based on the following equation:

$\begin{array}{cc}\mathrm{sign}<br><br>\left(k\right)=\{\begin{array}{cc}{\left(-1\right)}^k& \mathrm{if}<br><br><br><br>{\mathrm{\alpha <br><br>}}_1<<br><br>0\\ {\left(-1\right)}^{k+1}& \mathrm{if}<br><br><br><br>{\mathrm{\alpha <br><br>}}_1\ge <br><br>0\end{array},0<<br><br>k<<br><br>M,& \left(2\right)\end{array}$

wherein k is the channel number, and α

₁ is the first-order coefficient.

14. Apparatus in accordance with claim 10, in which the predetermined relationship is defined such that the audio subband signal having associated therewith the channel number k has a first sign and the adjacent audio subband signal having associated therewith the channel number k−

1 has a second sign, which is opposite to the first sign.

15. Apparatus in accordance with claim 14, in which the first sign is negative, and the second sign is positive.

16. Apparatus in accordance with claim 1, in which the examiner is operative to perform a tonal analysis for the audio subband signal and the adjacent audio subband signal for determining a tonal component having a tonality measure above a tonality threshold.

17. Apparatus in accordance with claim 16, in which the examiner is operative to determine, whether the tonal component is in the overlapping range of channel k and channel k−

1.

18. Apparatus in accordance with claim 1, further comprising a synthesis filterbank for filtering the gain adjusted audio subband signals to obtain a synthesized output signal.

19. Apparatus in accordance with claim 1, in which the analysis filterbank is a real valued filterbank, and in which the synthesis filterbank is a real valued filterbank.

20. Apparatus in accordance with claim 1, in which the analysis filterbank is a complex-valued filterbank, and in which the synthesis filterbank is a real-valued filterbank.

21. Method of calculating gain adjustment values for gain adjusting a plurality of audio subband signals generated by filtering a signal using an analysis filterbank, the filterbank having subband filters, adjacent subband filters of the filterbank having transition bands overlapping in an overlapping range, comprising:

examining an audio subband signal originating from a subband filter and an adjacent audio subband signal originating from an adjacent subband filter, to determine whether the audio subband signal and the adjacent audio subband signal have aliasing generating signal components in the overlapping range to obtain grouped audio subband signals in response to a result of the examination indicating that the audio subband signal and the adjacent audio subband signal have aliasing generating signal components in the overlapping range;

calculating a first gain adjustment value and a second gain adjustment value for the grouped adjacent audio subband signals,

by determining an indication of a reference energy for the grouped audio subband signals, an energy estimate for an energy in the grouped adjacent audio subband signals, and

by calculating the gain adjustment values for the grouped audio subband signals based on the indication of the reference energy and the energy estimate.

22. Digital storage medium having stored thereon a computer program having a program code for accessing an energy of a signal in accordance with claim 21, when the computer program runs on a computer.