SYNTHESIS OF AUDIO SCENARIOS BASED ON LATE REFLECTION

Abstract

A method of audio processing for synthesizing an auditory scene, comprising: processing (702) at least one input channel (312), using an auditory filter bank block (702), to generate two or more processed input signals (704); filtering (720) the at least one input channel (312), using a filter (720) that models late reverberation (LR), to generate corresponding two or more LR-filtered diffuse signals (722); and for each of the two or more processed input signals and each of the corresponding two or more diffuse signals, combining (714) one of the two or more LR-filtered diffuse signals with a corresponding one of the two or more processed input signals to generate one of a plurality of output channels (324) for the auditory scene, the method further comprising: converting (702) the at least one input channel (312) from a time domain into a frequericy domain to generate a plurality of frequency-domain (FD) input signals (704); and wherein processing (702) the at least one input channel (312) comprises: delaying (706) and scaling (710) the FD input signals to generate a plurality of scaled, delayed FD signals (712) as processed input signals, wherein each FD input signal is delayed at a corresponding delay block based on delay values derived from corresponding inter-channel time difference data, and wherein each delayed FD input signal is scaled by a corresponding multiplier based on scale factors derived from corresponding inter-channel level difference data.

Description

[0001] Regarding European patent 1 565 036 (DE 60 2005 053 100) The 5th Senate (Voidance Senate) of the Federal Patent Court, based on the oral hearing of April 17, 2023, has decided recognized as rightful: I. The European patent EP 1 565 036 is declared invalid with effect for the territory of the Federal Republic of Germany, subject to the condition that its patent claims shall be replaced by the following wording:

Claims

[1] A method of audio processing for synthesizing an auditory scene, comprising: processing (702) at least one input channel (312), using an auditory filter bank block (702), to generate two or more processed input signals (704); filtering (720) the at least one input channel (312), using a filter (720) that models late reverberation (LR), to generate corresponding two or more LR-filtered diffuse signals (722); and for each of the two or more processed input signals and each of the corresponding two or more diffuse signals, combining (714) one of the two or more LR-filtered diffuse signals with a corresponding one of the two or more processed input signals to generate one of a plurality of output channels (324) for the auditory scene, the method further comprising: converting (702) the at least one input channel (312) from a time domain into a frequericy domain to generate a plurality of frequency-domain (FD) input signals (704); and wherein processing (702) the at least one input channel (312) comprises: delaying (706) and scaling (710) the FD input signals to generate a plurality of scaled, delayed FD signals (712) as processed input signals, wherein each FD input signal is delayed at a corresponding delay block based on delay values derived from corresponding inter-channel time difference data, and wherein each delayed FD input signal is scaled by a corresponding multiplier based on scale factors derived from corresponding inter-channel level difference data. [2] The method of claim 1, wherein: the LR-filtered diffuse signals (722) are FD diffuse signals; and the combining (714) comprises, for each output channel: summing (714) one of the scaled, delayed FD signals (712) and a corresponding one of the FD diffuse signals (730) to generate an FD output signal (716); and converting (718) the FD output signal (716) from the frequency domain into the time domain to generate one of a plurality of output channels (324). [3] The method of claim 2, wherein filtering (720) the at least one input channel (312) comprises: applying two or more late reverberation filters (720) to the at least one input channel (312) to generate a plurality of LR-filtered diffuse signals (722); converting (724) the LR-filtered diffuse signals (712) from the time domain into the frequency domain to generate a plurality of FD diffuse signals (726); and scaling (728) the FD diffuse signals (726) to generate a plurality of scaled FD diffuse signals (730), wherein the scaled FD diffuse signals (730) are combined with the scaled, delayed FD signals (712) to generate the FD output signals (716). [4] The method of claim 1, wherein filtering the at least one input channel comprises: applying two or more FD late reverberation filters to the FD input signals to generate a plurality of diffuse FD signals; and scaling the diffuse FD signals to generate a plurality of scaled diffuse FD signals, wherein the scaled diffuse FD signals are combined with the scaled, delayed FD signals to generate an FD output signal. [5] The method of claim 1, wherein the method: applies the processing, filtering, and combining for input channel frequencies less than a specified threshold frequency; and further applies alternative auditory scene synthesis processing for input channel frequencies greater than the specified threshold frequency. [6] The method of claim 5, wherein the alternative auditory scene synthesis processing involves coherence-based binaural cue coding (BCC) without the filtering that is appliedto the input channel frequencies less than the specified threshold frequency. [7] Apparatus (322) for audio processing including synthesizing an auditory scene, comprising: means (702) for processing at least one input channel (312) to generate two or more processed input signals (704); means (720) for filtering the at least one input channel (312), using a filter that models late reverberation (LR) to generate corresponding two or more LR-filtered diffuse signals (722); and means (714) for combining, for each of the two or more processed input signals and each of the corresponding two or more diffuse signals, one of the two or more LR-filtered diffuse signals with a corresponding one of the two or more processed input signals to generate one of a plurality of output channels (324) for the auditory scene, wherein the means for processing the at least one input channel is configured to convert (702) the at least one input channel (312) from a time domain into a frequency domain to generate a plurality of frequency-domain (FD) input signals (704); and to delay (706) and scale (710) the FD input signals to generate a plurality of scaled, delayed FD signals (712) as processed input signals, wherein each FD input signal is delayed at a corresponding delay block based on delay values derived from corresponding inter-channel time difference data, and wherein each delayed FD input signal is scaled by a corresponding multiplier based on scale factors derived from corresponding inter-channel level difference data. [8] Apparatus (322) for audio processing including synthesizing an auditory scene, comprising: a configuration of at least one time domain (TD) to frequency domain (FD) converter (702) and a plurality of filters (720) that model late reverberation (LR), the configuration adapted to generate two or more processed FD input signals (704) and corresponding two or more LR-filtered diffuse FD signals (722) from at least one TD input channel (312), wherein each FD input signal is delayed at a corresponding delay block based on delay values derived from corresponding inter-channel time difference data, and wherein each delayed FD input signal is scaled by a corresponding multiplier based on scale factors derived from corresponding inter-channel level difference data; two or more combiners (714), each being adapted to combine one of the two or more LR-filtered diffuse FD signals (730) with a corresponding one of the two or more processed FD input signals (712) to generate a plurality of synthesized FD signals (716); and two or more frequency domain to time domain (FD-TD) converters (718), each adapted to convert one of the synthesized FD signals (716) into one of a plurality of TD output channels (324) for the auditory scene. [9] The apparatus of claim 8, wherein at least two filters (720) have different filter lengths.