Abstract:
A system and method is described for estimating residual echo and performing residual echo suppression on a signal. In particular, the system and method described herein: 1) provides a more accurate coherence-based approach where the sub-band nature of coherence-based residual echo estimation is leveraged using sub-band-based coherence measures and upper-bound envelopes to guide residual echo estimates; 2) provides joint estimation of the residual echo and a background stationary noise-floor; 3) provides more accurate residual echo estimation by combining the above two measures with information on near-end activity to selectively adjust the residual-echo estimation in sub-bands; and 4) provides improved control of near-end distortion by leveraging the noise-floor estimate and the internally calculated signal-to-echo ratios as additional guides/limits to help balance, flexibly, and as needed, near-end distortion with residual echo suppression. The resultant residual echo suppressor design provides significant improvement in both residual echo suppression and background noise preservation.
Abstract:
A device implementing a system for mitigating noise includes at least one processor configured to receive a first audio signal corresponding to a first microphone, and determine whether wind noise is present based at least in part on the first audio signal. The processor is configured to select, based on the determining, a second audio signal from between second and third microphones. The second microphone is disposed at a location that experiences less echo coupling when the device is in a particular orientation with respect to a user. The third microphone is disposed at another location that experiences less wind noise. The processor is configured to determine voice and noise reference values based on the first and the selected second audio signals, and perform noise suppression with respect to at least one of the first or the selected second audio signal, based on the voice or the noise reference value.
Abstract:
To address issues with present echo gate control, a method and apparatus for more intelligently operating an echo gate is described herein. In particular, the decision of whether to mute an uplink signal, or not, is formulated herein as primarily a perceptual decision based on an appropriate analysis of the perceptual interaction of the current residual echo and the current near-end signal(s). By doing so, the application of muting through an echo gate may be minimized and/or more appropriately engaged. This will lead to fewer dropouts and muting of speech onsets and offsets 1) during periods such as double-talk or 2) during periods of downlink playback in the presence of low near-end signal levels, two cases of particular importance.
Abstract:
A system and method is described for estimating residual echo and performing residual echo suppression on a signal. In particular, the system and method described herein: 1) provides a more accurate coherence-based approach where the sub-band nature of coherence-based residual echo estimation is leveraged using sub-band-based coherence measures and upper-bound envelopes to guide residual echo estimates; 2) provides joint estimation of the residual echo and a background stationary noise-floor; 3) provides more accurate residual echo estimation by combining the above two measures with information on near-end activity to selectively adjust the residual-echo estimation in sub-bands; and 4) provides improved control of near-end distortion by leveraging the noise-floor estimate and the internally calculated signal-to-echo ratios as additional guides/limits to help balance, flexibly, and as needed, near-end distortion with residual echo suppression. The resultant residual echo suppressor design provides significant improvement in both residual echo suppression and background noise preservation.
Abstract:
To address issues with present echo gate control, a method and apparatus for more intelligently operating an echo gate is described herein. In particular, the decision of whether to mute an uplink signal, or not, is formulated herein as primarily a perceptual decision based on an appropriate analysis of the perceptual interaction of the current residual echo and the current near-end signal(s). By doing so, the application of muting through an echo gate may be minimized and/or more appropriately engaged. This will lead to fewer dropouts and muting of speech onsets and offsets 1) during periods such as double-talk or 2) during periods of downlink playback in the presence of low near-end signal levels, two cases of particular importance.