VoiceFilter-Lite: Streaming Targeted Voice Separation for On-Device Speech Recognition

Abstract

We introduce VoiceFilter-Lite, a single-channel source separation model that runs on the device to preserve only the speech signals from a target user, as part of a streaming speech recognition system. Delivering such a model presents numerous challenges: It should improve the performance when the input signal consists of overlapped speech, and must not hurt the speech recognition performance under all other acoustic conditions. Besides, this model must be tiny, fast, and perform inference in a streaming fashion, in order to have minimal impact on CPU, memory, battery and latency. We propose novel techniques to meet these multi-faceted requirements, including using a new asymmetric loss, and adopting adaptive runtime suppression strength. We also show that such a model can be quantized as a 8-bit integer model and run in realtime.

References in corpus (7)

• SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
• Sequence Transduction with Recurrent Neural Networks
• Deep attractor network for single-microphone speaker separation
• Optimizing Speech Recognition For The Edge
• Deep Extractor Network for Target Speaker Recovery From Single Channel Speech Mixtures
• Version Control of Speaker Recognition Systems
• On the efficient representation and execution of deep acoustic models

Cited by in corpus (2)

• Improving RNN Transducer With Target Speaker Extraction and Neural Uncertainty Estimation
• Investigation of Practical Aspects of Single Channel Speech Separation for ASR