CyclicAugment: Speech Data Random Augmentation with Cosine Annealing Scheduler for Automatic Speech Recognition

Zhihan Wang, Feng Hou, Yuanhang Qiu, Zhizhong Ma, Satwinder Singh, Ruili Wang

Research output: Journal PublicationConference articlepeer-review

4 Citations (Scopus)

Abstract

Recent speech data augmentation approaches use static augmentation operations or policies with consistency magnitude scaling. However, few work is done to explore the influence of the dynamic magnitude of augmentation policies. In this paper, we propose a novel speech data augmentation approach, CyclicAugment, to generate more diversified augmentation policies by dynamically configuring the magnitude of augmentation policies with a cosine annealing scheduler. We also propose additional augmentation operations to enlarge the diversity of augmentation policies. Motivated by learning rate warm restart and cyclical learning rates, we hypothesize that using dynamically configured magnitude for augmentation policies can also help escape local optima more efficiently than static augmentation policies with consistency magnitude scaling. Experimental results demonstrate that our approach is effective for escaping local optima. Our approach achieves 12%-35% relative improvement in word error rate (WER) over SpecAugment and RandAugment on the LibriSpeech 960h dataset, and achieves state-of-the-art result 7.1% in phoneme error rate (PER) on the TIMIT 5h dataset.

Original languageEnglish
Pages (from-to)3859-3863
Number of pages5
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Volume2022-September
DOIs
Publication statusPublished - 2022
Externally publishedYes
Event23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022 - Incheon, Korea, Republic of
Duration: 18 Sept 202222 Sept 2022

Keywords

  • cosine annealing scheduler
  • data augmentation
  • random augmentation
  • speech recognition

ASJC Scopus subject areas

  • Language and Linguistics
  • Human-Computer Interaction
  • Signal Processing
  • Software
  • Modelling and Simulation

Fingerprint

Dive into the research topics of 'CyclicAugment: Speech Data Random Augmentation with Cosine Annealing Scheduler for Automatic Speech Recognition'. Together they form a unique fingerprint.

Cite this