Audio gain control using specific-loudness-based auditory event detection
In one disclosed aspect, dynamic gain modifications are applied to an audio signal at least partly in response to auditory events and/or the degree of change in signal characteristics associated with said auditory event boundaries. In another aspect, an audio signal is divided into auditory events by comparing the difference in specific loudness between successive time blocks of the audio signal.
1. An audio processing method to generate a second signal by applying dynamic gain modifications to a first signal, the method comprising:
detecting changes in loudness with respect to time in the first signal,
identifying as auditory event boundaries changes greater than a threshold in loudness with respect to time in the first signal, wherein an audio segment between consecutive boundaries constitutes an auditory event, and
applying a dynamic gain modification to the first signal based on said auditory event to generate the second signal;
wherein the audio processing method is performed by one or more processors.
2. The method according to claim 1 wherein the detecting includes comparing changes in loudness with respect to time in the first signal across frequency bands.
3. The method according to claim 1 wherein the applying a dynamic gain modification modifies a gain-smoothing time constant.
4. The method according to claim 3 wherein the applying a dynamic gain modification modifies a gain-smoothing attack time constant.
5. The method according to claim 3 wherein the applying a dynamic gain modification modifies a gain-smoothing release time constant.
6. The method according to claim 1 wherein loudness is perceptual loudness.
7. The method according to claim 6 wherein the applying a dynamic gain modification modifies a gain-smoothing time constant.
8. The method according to claim 7 wherein the applying a dynamic gain modification modifies a gain-smoothing attack time constant.
9. The method according to claim 7 wherein the applying a dynamic gain modification modifies a gain-smoothing release time constant.
10. The method according to claim 1 wherein the auditory event is a voice segment.
11. The method according to claim 10 wherein the applying a dynamic gain modification modifies a gain-smoothing time constant.
12. A non-transitory computer-readable storage medium encoded with a computer program for causing a computer to perform the method of claim 1 .
13. An audio processing method in which a processor receives an input channel and generates an output channel that is generated by applying a dynamic gain modification to the input channel, comprising
detecting changes in loudness with respect to time in the audio input channel,
identifying as auditory event boundaries changes in loudness greater than a threshold with respect to time in said input channel, wherein an audio segment between consecutive boundaries constitutes an auditory event in the input channel, and
generating one or more dynamically-varying parameters in response to the auditory events and/or the degree of change in loudness associated with said auditory event boundaries;
wherein the audio processing method is performed by one or more processors.