Personal VAD: Speaker-Conditioned Voice Activity Detection

Citations Over TimeTop 10% of 2020 papers

Abstract

In this paper, we propose "personal VAD", a system to detect the voice activity of a target speaker at the frame level.This system is useful for gating the inputs to a streaming on-device speech recognition system, such that it only triggers for the target user, which helps reduce the computational cost and battery consumption, especially in scenarios where a keyword detector is unpreferable.We achieve this by training a VAD-alike neural network that is conditioned on the target speaker embedding or the speaker verification score.For each frame, personal VAD outputs the probabilities for three classes: non-speech, target speaker speech, and non-target speaker speech.Under our optimal setup, we are able to train a model with only 130K parameters that outperforms a baseline system where individually trained standard VAD and speaker recognition networks are combined to perform the same task.

Related Papers

• A Robust Voice Activity Detection for Real-Time Automatic Speech Recognition(2018)
• A Review on Speech Recognition Methods(2018)
• → Multi-speaker adaptation for robust speech recognition under ubiquitous environment(2009)1 cited
• → End-to-End Speaker-Dependent Voice Activity Detection(2020)2 cited
• A datasieve based audiovisual continuous speech recognition system(2003)