Improving sequence-to-sequence speech recognition training with on-the-fly data augmentation

10/29/2019
by   Thai-Son Nguyen, et al.
57

Sequence-to-Sequence (S2S) models recently started to show state-of-the-art performance for automatic speech recognition (ASR). With these large and deep models overfitting remains the largest problem, outweighing performance improvements that can be obtained from better architectures. One solution to the overfitting problem is increasing the amount of available training data and the variety exhibited by the training data with the help of data augmentation. In this paper we examine the influence of three data augmentation methods on the performance of two S2S model architectures. One of the data augmentation method comes from literature, while two other methods are our own development - a time perturbation in the frequency domain and sub-sequence sampling. Our experiments on Switchboard and Fisher data show state-of-the-art performance for S2S models that are trained solely on the speech training data and do not use additional text data.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/03/2021

On-the-Fly Aligned Data Augmentation for Sequence-to-Sequence ASR

We propose an on-the-fly data augmentation method for automatic speech r...
research
08/16/2023

Quantifying Overfitting: Introducing the Overfitting Index

In the rapidly evolving domain of machine learning, ensuring model gener...
research
04/27/2022

Improving Multimodal Speech Recognition by Data Augmentation and Speech Representations

Multimodal speech recognition aims to improve the performance of automat...
research
01/20/2020

Single headed attention based sequence-to-sequence model for state-of-the-art results on Switchboard-300

It is generally believed that direct sequence-to-sequence (seq2seq) spee...
research
04/21/2021

Disfluency Detection with Unlabeled Data and Small BERT Models

Disfluency detection models now approach high accuracy on English text. ...
research
04/22/2022

Efficient Training of Neural Transducer for Speech Recognition

As one of the most popular sequence-to-sequence modeling approaches for ...
research
05/20/2018

Generating High-Quality Surface Realizations Using Data Augmentation and Factored Sequence Models

This work presents a new state of the art in reconstruction of surface r...

Please sign up or login with your details

Forgot password? Click here to reset