Voice activity detection in presence of background noise

Abstract

In speech processing systems, compensation is made for sudden changes in the background noise in the average signal-to-noise ratio (SNR) calculation. SNR outlier filtering may be used, alone or in conjunction with weighting the average SNR. Adaptive weights may be applied on the SNRs per band before computing the average SNR. The weighting function can be a function of noise level, noise type, and / or instantaneous SNR value. Another weighting mechanism applies a null filtering or outlier filtering which sets the weight in a particular band to be zero. This particular band may be characterized as the one that exhibits an SNR that is several times higher than the SNRs in other bands.

Description

[0001] Cross References to Related Applications [0002] This application claims priority to Provisional Patent Application No. 61 / 588,729, filed January 20, 2012, on the basis of 35 U.S.C. § 119(e). This provisional patent application is expressly incorporated herein by reference in its entirety. Background technique [0003] For applications where communications occur in noisy environments, it may be desirable to separate the desired speech signal from background noise. Noise may be defined as the combination of all signals that interfere with or otherwise degrade a desired signal. Background noise may include many noise signals generated within an acoustic environment, such as background conversations of other people, as well as reflections and reverberations from any of the desired signal and / or other signals. [0004] A signal activity detector, such as a voice activity detector (VAD), can be used to minimize the amount of unnecessary processing in the electronic devic...

AI Technical Summary

Problems solved by technology

At low SNR or under rapidly changing non-stationary noise, a smooth long-term SNR will produce an inaccurate VAD_THR, resulting in an increased probability of missed speech or an increased probability of false speech detection

Also, some VAD techniques (e.g., Adaptive Multi-Rate Wideband or AMR-WB) work well for stationary types of noise, such as car noise, but produce extremely high Voice activity factor (attributable to widespread false detections)

[0008] Therefore, false indications of voice activity can lead to processing and emission of noisy signals

The processing and emission of noisy signals can create a poor user experience, especially if the noise emission period is interrupted by periods of inactivity from time to time due to voice activity detectors indicating no voice activity

Conversely, poor voice activity detection can lead to loss of a substantial portion of the voice signal

Loss of the initial part of the voice activity can result in the user needing to repeat parts of the conversation on a regular basis, which is an undesirable situation

Images