Synthesis Speech Based Data Augmentation for Low Resource Children ASR

Virender Kadyan, Hemant Kathania*, Prajjval Govil, Mikko Kurimo

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

21 Downloads (Pure)

Abstract

Successful speech recognition for children requires large training data with sufficient speaker variability. The collection of such a training database of children’s voices is challenging and very expensive for zero/low resource language like Punjabi. In this paper, the data scarcity issue of the low resourced language Punjabi is addressed through two levels of augmentation. The original training corpus is first augmented by modifying the prosody parameters for pitch and speaking rate. Our results show that the augmentation improves the system performance over the baseline system. Then the augmented data combined with original data and used to train the TTS system to generate synthesis data and extended dataset is further used for augmented by generating children’s utterances using text-to-speech synthesis and sampling the language model with methods that increase the acoustic and lexical diversity. The final speech recognition performance indicates a relative improvement of 50.10% with acoustic and 57.40% with language diversity based augmentation in comparison to that of the baseline system respectively.

Original languageEnglish
Title of host publicationSpeech and Computer - 23rd International Conference, SPECOM 2021, Proceedings
EditorsAlexey Karpov, Rodmonga Potapova
Pages317-326
Number of pages10
DOIs
Publication statusPublished - 2021
MoE publication typeA4 Article in a conference publication
EventInternational Conference on Speech and Computer - Virtual, Online
Duration: 27 Sep 202130 Sep 2021
Conference number: 23

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12997 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceInternational Conference on Speech and Computer
Abbreviated titleSPECOM
CityVirtual, Online
Period27/09/202130/09/2021

Keywords

  • Children speech recognition
  • Low resource
  • Prosody modification
  • Speech synthesis
  • Tacotron

Fingerprint

Dive into the research topics of 'Synthesis Speech Based Data Augmentation for Low Resource Children ASR'. Together they form a unique fingerprint.

Cite this