Method for generating filter for audio signal, and parameterization device for same

• US 10,701,511 B2
• Filed: 08/19/2019
• Issued: 06/30/2020
• Est. Priority Date: 12/23/2013
• Status: Active Grant

• Current Assignee: GCOA Company Limited, WILUS Inc.
• Original Assignee: WILUS Inc.
• Inventors: Lee, Taegyu, Oh, Hyunoh
• Primary Examiner(s): Laekemariam, Yosef K
• Application Number:
  US16/544,832
• Publication Number:
  US 20190373399A1
• Time in Patent Office: 316 Days
• Field of Search: 381 1, 381 17, 381 18, 381 23, 381 61, 381 63, 381306, 381307, 381309
• CPC Class Codes: G10L 19/008 Multichannel audio signal coding or decoding using interchannel correlation to reduce redundancy, e.g. joint-stereo, intensity-coding or matrixing
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: H04S 2400/01 Multi-channel, i.e. more than two input channels, sound reproduction with two speakers wherein the multi-channel information is substantially preserved
• CPC Class Codes: H04S 2400/03 Aspects of down-mixing multi-channel audio to configurations with lower numbers of playback channels, e.g. 7.1 -> 5.1 H04S2400/01 takes precedence
• CPC Class Codes: H04S 2420/01 Enhancing the perception of the sound image or of the spatial distribution using head related transfer functions [HRTF's] or equivalents thereof, e.g. interaural time difference [ITD] or interaural level difference [ILD]
• CPC Class Codes: H04S 2420/03 Application of parametric coding in stereophonic audio systems
• CPC Class Codes: H04S 2420/07 Synergistic effects of band splitting and sub-band processing
• CPC Class Codes: H04S 3/008 in which the audio signals are in digital form, i.e. employing more than two discrete digital channels data reduction aspects thereof based on psychoacoustics G10L19/02
• CPC Class Codes: H04S 7/305 Electronic adaptation of stereophonic audio signals to reverberation of the listening space H04S7/301 takes precedence
• CPC Class Codes: H04S 7/307 Frequency adjustment, e.g. tone control H04S7/301 takes precedence

Abstract

The present invention relates to a method for generating a filter for an audio signal and a parameterization device for the same, and more particularly, to a method for generating a filter for an audio signal, to implement filtering of an input audio signal with a low computational complexity, and a parameterization device therefor.

To this end, provided are a method for generating a filter for an audio signal, including: receiving at least one binaural room impulse response (BRIR) filter coefficients for binaural filtering of an input audio signal; converting the BRIR filter coefficients into a plurality of subband filter coefficients; obtaining average reverberation time information of a corresponding subband by using reverberation time information extracted from the subband filter coefficients; obtaining at least one coefficient for curve fitting of the obtained average reverberation time information; obtaining flag information indicating whether the length of the BRIR filter coefficients in a time domain is more than a predetermined value; obtaining filter order information for determining a truncation length of the subband filter coefficients, the filter order information being obtained by using the average reverberation time information or the at least one coefficient according to the obtained flag information and the filter order information of at least one subband being different from filter order information of another subband; and truncating the subband filter coefficient by using the obtained filter order information and a parameterization device therefor.

Claims

1. A method for generating a set of filter coefficients for an audio signal, comprising:

obtaining a set of subband filter coefficients for each subband, wherein the set of subband filter coefficients is obtained from a set of time domain binaural room impulse response (BRIR) filter coefficients;

obtaining average reverberation time information of each subband by using a reverberation time extracted from the set of subband filter coefficients for a corresponding subband;

obtaining one or more coefficients for curve fitting the average reverberation time information;

obtaining flag information indicating whether a length of the set of time domain BRIR filter coefficients is larger than a predetermined value;

determining a filter order of each subband, wherein the filter order is determined, based at least in part on the flag information, by using the average reverberation time information or by using the coefficients for curve fitting, and wherein the filter order is determined to be variable in a frequency domain; and

truncating the set of subband filter coefficients by using the filter order of a corresponding subband.

2. The method of claim 1, wherein when the flag information indicates that the length of the set of time domain BRIR filter coefficients is larger than the predetermined value, the filter order is determined based on a curve-fitted value which is obtained by using the coefficients for curve fitting.

3. The method of claim 2, wherein the curve-fitted value is a value of power of 2 having an approximated integer value as an index, and

wherein the approximated integer value is obtained by performing a polynomial curve-fitting by using the coefficients for curve fitting.

4. The method of claim 1, wherein when the flag information indicates that the length of the set of time domain BRIR filter coefficients is not larger than the predetermined value, the filter order is determined based on the average reverberation time information of a corresponding subband without performing the curve fitting.

5. The method of claim 4, wherein the filter order is determined to be a value of power of 2 having a log-scaled approximated integer value of the average reverberation time information as an index.

6. The method of claim 1, wherein the filter order is determined as a smaller value between a reference truncation length of a corresponding subband and an original length of the set of subband filter coefficients, and

wherein the reference truncation length of the corresponding subband is obtained by using the average reverberation time information or the coefficients for curve fitting.

7. The method of claim 6, wherein the reference truncation length is a value of power of 2.

8. The method of claim 1, wherein the filter order has a single value for each subband.

9. The method of claim 1, wherein the average reverberation time information is an average value of reverberation times of extracted respectively from a set of subband filter coefficients for each channel in the corresponding subband.

10. A parameterization device for generating a set of filter coefficients for an audio signal, the parameterization device configured to:

obtain a set of subband filter coefficients for each subband, wherein the set of subband filter coefficients is obtained from a set of time domain binaural room impulse response (BRIR) filter coefficients;

obtain average reverberation time information of each subband by using a reverberation time extracted from the set of subband filter coefficients for a corresponding subband;

obtain one or more coefficients for curve fitting the average reverberation time information;

obtain flag information indicating whether a length of the set of time domain BRIR filter coefficients is larger than a predetermined value;

determine a filter order of each subband, wherein the filter order is determined, based at least in part on the flag information, by using the average reverberation time information or by using the coefficients for curve fitting, and wherein the filter order is determined to be variable in a frequency domain; and

truncate the set of subband filter coefficients by using the filter order of a corresponding subband.

11. The parameterization device of claim 10, wherein when the flag information indicates that the length of the set of time domain BRIR filter coefficients is larger than the predetermined value, the filter order is determined based on a curve-fitted value which is obtained by using the coefficients for curve fitting.

12. The parameterization device of claim 11, wherein the curve-fitted value is a value of power of 2 having an approximated integer value as an index, and

wherein the approximated integer value is obtained by performing a polynomial curve-fitting by using the coefficients for curve fitting.

13. The parameterization device of claim 10, wherein when the flag information indicates that the length of the set of time domain BRIR filter coefficients is not larger than the predetermined value, the filter order is determined based on the average reverberation time information of a corresponding subband without performing the curve fitting.

14. The parameterization device of claim 13, wherein the filter order is determined to be a value of power of 2 having a log-scaled approximated integer value of the average reverberation time information as an index.

15. The parameterization device of claim 10, wherein the filter order is determined as a smaller value between a reference truncation length of a corresponding subband and an original length of the set of subband filter coefficients, and

wherein the reference truncation length of the corresponding subband is obtained by using the average reverberation time information or the coefficients for curve fitting.

16. The parameterization device of claim 15, wherein the reference truncation length is a value of power of 2.

17. The parameterization device of claim 10, wherein the filter order has a single value for each subband.

18. The parameterization device of claim 10, wherein the average reverberation time information is an average value of reverberation times extracted respectively from a set of subband filter coefficients for each channel in the corresponding subband.