Processing of an audio input signal

• US 10,638,227 B2
• Filed: 12/02/2016
• Issued: 04/28/2020
• Est. Priority Date: 12/02/2016
• Status: Active Grant

• Current Assignee: Dirac Research AB
• Original Assignee: Dirac Research AB
• Inventors: Arlbrant, Mattias
• Primary Examiner(s): Matar, Ahmad F.
• Assistant Examiner(s): Diaz, Sabrina
• Application Number:
  US16/466,096
• Publication Number:
  US 20190306624A1
• Time in Patent Office: 1,243 Days
• Field of Search: 381 55, 381 941, 381 942, 381 98
• CPC Class Codes: G10L 21/02 Speech enhancement, e.g. noise reduction or echo cancellation reducing echo effects in line transmission systems H04B3/20; echo suppression in hands-free telephones H04M9/08
• CPC Class Codes: G10L 21/0324 Details of processing therefor
• CPC Class Codes: H03G 3/3005 in amplifiers suitable for low-frequencies, e.g. audio amplifiers H03G3/32, H03G3/34 take precedence
• CPC Class Codes: H03G 9/005 of digital or coded signals
• CPC Class Codes: H03G 9/025 frequency-dependent volume compression or expansion, e.g. multiple-band systems H03G9/10, H03G9/18 take precedence
• CPC Class Codes: H04R 2430/01 Aspects of volume control, not necessarily automatic, in sound systems
• CPC Class Codes: H04R 2430/03 Synergistic effects of band splitting and sub-band processing
• CPC Class Codes: H04R 2499/11 Transducers incorporated or for use in hand-held devices, e.g. mobile phones, PDA's, camera's
• CPC Class Codes: H04R 3/007 Protection circuits for transducers
• CPC Class Codes: H04R 3/08 of electromagnetic transducers

Abstract

There is provided a method for processing an audio input signal and a corresponding audio filter system. The method includes applying non-linear time-domain processing to the input signal to generate a processed copy of the input signal, transforming the input signal and the processed copy of the input signal to the frequency domain, and performing a comparison in the frequency-domain based on the transformed input signal and the transformed processed copy of the input signal. The method also includes determining at least one control parameter based on the comparison, performing frequency-domain processing of the transformed input signal based on the control parameter, and generating an output signal by transforming the frequency-domain processed signal to the time domain.

Claims

1. A method for processing an audio input signal, the method comprising:

applying non-linear time-domain processing to the input signal to generate a processed copy of the input signal;

transforming the input signal and the processed copy of the input signal to the frequency domain;

performing a comparison in the frequency-domain based on the transformed input signal and the transformed processed copy of the input signal;

determining at least one control parameter based on said comparison;

performing frequency-domain processing of the transformed input signal based on said at least one control parameter; and

generating an output signal by transforming the frequency-domain processed transformed input signal to the time domain.

2. The method of claim 1, wherein characteristics present in the processed copy of the input signal are enhanced or suppressed.

3. The method of claim 1, wherein said frequency-domain processing suppresses frequency components of the transformed input signal such that levels of the frequency components do not exceed the levels of corresponding frequency components in the transformed processed copy of the input signal.

4. The method of claim 1, wherein said at least one control parameter includes a frequency-dependent gain, and said frequency-domain processing includes multiplication of the frequency-domain representation of the transformed input signal with said gain.

5. The method of claim 4, wherein a maximum attenuation of said gain is limited to a frequency-dependent threshold, and a frequency-dependent spectral smoothing is applied to said gain.

6. The method of claim 1, wherein the comparison comprises selecting the minimum of a predetermined value and a ratio between a magnitude of the transformed input signal and a magnitude of the transformed processed copy of the input signal.

7. The method of claim 1, wherein the non-linear time-domain processing of the input signal comprises clipping of the input signal to a predetermined level.

8. The method of claim 1, wherein the non-linear time-domain processing of the input signal comprises processing corresponding to clipping a membrane excursion of an electrodynamic loudspeaker to a predetermined level.

9. An audio filter system configured to process an audio input signal, the audio filter system comprising:

at least one processor configured to

apply non-linear time-domain processing to the input signal to generate a processed copy of the input signal,

transform the input signal and the processed copy of the input signal to the frequency domain,

perform a comparison in the frequency-domain based on the transformed input signal and the transformed processed copy of the input signal,

determine at least one control parameter based on said comparison,

perform frequency-domain processing of the transformed input signal based on said at least one control parameter, and

generate an output signal by transforming the frequency-domain processed transformed input signal to the time domain.

10. The audio filter system of claim 9, wherein said at least one processor is configured to enhance or suppress characteristics present in the processed copy of the input signal.

11. The audio filter system of claim 9, wherein said at least one processor is configured to suppress frequency components of the transformed input signal such that levels of the frequency components do not exceed the levels of corresponding frequency components in the transformed processed copy of the input signal.

12. The audio filter system of claim 9, wherein said at least one control parameter includes a frequency-dependent gain, and said frequency-domain processing includes multiplication of the frequency-domain representation of the transformed input signal with said gain.

13. The audio filter system of claim 12, wherein said at least one processor is configured to limit a maximum attenuation of said gain to a frequency-dependent threshold, and to apply a frequency-dependent spectral smoothing to said gain.

14. The audio filter system of claim 9, wherein the comparison comprises selecting the minimum of a predetermined value and a ratio between a magnitude of the transformed input signal and a magnitude of the transformed processed copy of the input signal.

15. The audio filter system of claim 9, wherein the non-linear time-domain processing of the input signal comprises clipping of the input signal to a predetermined level.

16. The audio filter system of claim 9, wherein the non-linear time-domain processing of the input signal comprises processing corresponding to clipping a membrane excursion of an electrodynamic loudspeaker to a predetermined level.

17. The audio filter system of claim 9, further comprising a memory comprising instructions, which when executed by the at least one processor, cause the at least one processor to process the audio input signal.

18. The audio filter system of claim 9, wherein the audio filter system comprises at least one post-processing stage to perform additional signal processing.

20. An audio system comprising:

the audio filter system of claim 9.

19. A non-transitory computer-readable medium having stored thereon a computer program for processing, when executed by a processor, an audio input signal, wherein said computer program comprises instructions, which when executed by the processor, cause the processor to:

apply non-linear time-domain processing to the input signal to generate a processed copy of the input signal;

transform the input signal and the processed copy of the input signal to the frequency domain;

perform a comparison in the frequency-domain based on the transformed input signal and the transformed processed copy of the input signal;

determine at least one control parameter based on said comparison;

perform frequency-domain processing of the transformed input signal based on said at least one control parameter; and

generate an output signal by transforming the frequency-domain processed transformed input signal to the time domain.