Incorporating VAD into ASR System by Multi-task Learning

TL;DR

This paper introduces a multi-task learning framework that integrates voice activity detection into end-to-end speech recognition, enhancing accuracy and efficiency by jointly training both tasks and leveraging VAD information.

Contribution

The paper presents a novel multi-task learning approach that incorporates VAD into ASR, improving performance and robustness over traditional independent systems.

Findings

Outperforms baseline ASR on English and Chinese datasets.

Better handles unsegmented speech data compared to GMM/DNN-based VAD systems.

Reduces computational cost by discarding non-speech parts during inference.

Abstract

When we use End-to-end automatic speech recognition (E2E-ASR) system for real-world applications, a voice activity detection (VAD) system is usually needed to improve the performance and to reduce the computational cost by discarding non-speech parts in the audio. Usually ASR and VAD systems are trained and utilized independently to each other. In this paper, we present a novel multi-task learning (MTL) framework that incorporates VAD into the ASR system. The proposed system learns ASR and VAD jointly in the training stage. With the assistance of VAD, the ASR performance improves as its connectionist temporal classification (CTC) loss function can leverage the VAD alignment information. In the inference stage, the proposed system removes non-speech parts at low computational cost and recognizes speech parts with high robustness. Experimental results on segmented speech data show that by utilizing VAD information, the proposed method outperforms the baseline ASR system on both English and Chinese datasets. On unsegmented speech data, we find that the system outperforms the ASR systems that build an extra GMM-based or DNN-based voice activity detector.