LEAP Submission to CHiME-6 ASR Challenge}

TL;DR

This paper presents LEAP's ASR system for the CHiME-6 challenge, utilizing data augmentation and advanced neural architectures to improve speech recognition in challenging noisy home environments.

Contribution

The paper introduces a novel combination of data augmentation and a hybrid TDNN-LSTM neural architecture for robust speech recognition in noisy conditions.

Findings

2% relative WER improvement over baseline

Effective use of data augmentation techniques

Hybrid TDNN-LSTM architecture enhances recognition accuracy

Abstract

This paper reports the LEAP submission to the CHiME-6 challenge. The CHiME-6 Automatic Speech Recognition (ASR) challenge Track 1 involved the recognition of speech in noisy and reverberant acoustic conditions in home environments with multiple-party interactions. For the challenge submission, the LEAP system used extensive data augmentation and a factorized time-delay neural network (TDNN) architecture. We also explored a neural architecture that interleaved the TDNN layers with LSTM layers. The submitted system improved the Kaldi recipe by 2% in terms of relative word-error-rate improvements.