The JHU Multi-Microphone Multi-Speaker ASR System for the CHiME-6 Challenge

TL;DR

This paper presents the JHU team's multi-microphone speech recognition system for the CHiME-6 challenge, utilizing advanced multi-array processing techniques to improve diarization and recognition in home environments.

Contribution

The paper introduces novel multi-array processing methods and integrates various techniques to enhance speech diarization and recognition in challenging real-world settings.

Findings

Achieved WER of 40.5% and 67.5% on tracks 1 and 2, respectively.

Improved over baseline by 10.8% and 10.4% absolute WER.

Demonstrated effectiveness of multi-array processing and dereverberation techniques.

Abstract

This paper summarizes the JHU team's efforts in tracks 1 and 2 of the CHiME-6 challenge for distant multi-microphone conversational speech diarization and recognition in everyday home environments. We explore multi-array processing techniques at each stage of the pipeline, such as multi-array guided source separation (GSS) for enhancement and acoustic model training data, posterior fusion for speech activity detection, PLDA score fusion for diarization, and lattice combination for automatic speech recognition (ASR). We also report results with different acoustic model architectures, and integrate other techniques such as online multi-channel weighted prediction error (WPE) dereverberation and variational Bayes-hidden Markov model (VB-HMM) based overlap assignment to deal with reverberation and overlapping speakers, respectively. As a result of these efforts, our ASR systems achieve a word error rate of 40.5% and 67.5% on tracks 1 and 2, respectively, on the evaluation set. This is an improvement of 10.8% and 10.4% absolute, over the challenge baselines for the respective tracks.