Data Augmentation Methods for End-to-end Speech Recognition on Distant-Talk Scenarios

by   Emiru Tsunoo, et al.

Although end-to-end automatic speech recognition (E2E ASR) has achieved great performance in tasks that have numerous paired data, it is still challenging to make E2E ASR robust against noisy and low-resource conditions. In this study, we investigated data augmentation methods for E2E ASR in distant-talk scenarios. E2E ASR models are trained on the series of CHiME challenge datasets, which are suitable tasks for studying robustness against noisy and spontaneous speech. We propose to use three augmentation methods and thier combinations: 1) data augmentation using text-to-speech (TTS) data, 2) cycle-consistent generative adversarial network (Cycle-GAN) augmentation trained to map two different audio characteristics, the one of clean speech and of noisy recordings, to match the testing condition, and 3) pseudo-label augmentation provided by the pretrained ASR module for smoothing label distributions. Experimental results using the CHiME-6/CHiME-4 datasets show that each augmentation method individually improves the accuracy on top of the conventional SpecAugment; further improvements are obtained by combining these approaches. We achieved 4.3% word error rate (WER) reduction, which was more significant than that of the SpecAugment, when we combine all three augmentations for the CHiME-6 task.



There are no comments yet.


page 1

page 2

page 3

page 4


MixSpeech: Data Augmentation for Low-resource Automatic Speech Recognition

In this paper, we propose MixSpeech, a simple yet effective data augment...

You Do Not Need More Data: Improving End-To-End Speech Recognition by Text-To-Speech Data Augmentation

Data augmentation is one of the most effective ways to make end-to-end a...

A Cycle-GAN Approach to Model Natural Perturbations in Speech for ASR Applications

Naturally introduced perturbations in audio signal, caused by emotional ...

Noisy Training Improves E2E ASR for the Edge

Automatic speech recognition (ASR) has become increasingly ubiquitous on...

Multi-style Training for South African Call Centre Audio

Mismatched data is a challenging problem for automatic speech recognitio...

Text Augmentation for Language Models in High Error Recognition Scenario

We examine the effect of data augmentation for training of language mode...

Improving Multimodal Speech Recognition by Data Augmentation and Speech Representations

Multimodal speech recognition aims to improve the performance of automat...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.