Non-native speech recognition using audio style transfer

Kacper Radzikowski, Mateusz Forc, Le Wang, Osamu Yoshie, Robert M. Nowak

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Recently automatic speech recognition (ASR) systems achieve higher and higher accuracy rates. However, the score drops significantly, when the ASR system is being used with a non-native speaker of the language to be recognized, mainly because of specific pronunciation and accent features. A limited volume of labeled datasets containing samples of a non-native speech makes it difficult to train any new ASR systems targeted for non-native speakers. In our research, we tried tackling the problem of a non-native accent and its influence on the accuracy of ASR systems, using the style transfer methodology. We designed a pipeline for modifying the speech produced by a nonnative speaker, so that it resembles the native speech to a higher extent, i.e. a method for accent neutralization. Our methodology can be used as a wrapper for any existing ASR system, which reduces the necessity of training new speech recognizers, adapted for non-native speech. The modification can be thus performed on the fly, before passing the data forward to the speech recognition system itself.

Original languageEnglish
Title of host publicationPhotonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2019
EditorsRyszard S. Romaniuk, Maciej Linczuk
PublisherSPIE
ISBN (Electronic)9781510630659
DOIs
Publication statusPublished - 2019
Externally publishedYes
EventPhotonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2019 - Wilga, Poland
Duration: 2019 May 262019 Jun 2

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11176
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferencePhotonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2019
CountryPoland
CityWilga
Period19/5/2619/6/2

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Non-native speech recognition using audio style transfer'. Together they form a unique fingerprint.

  • Cite this

    Radzikowski, K., Forc, M., Wang, L., Yoshie, O., & Nowak, R. M. (2019). Non-native speech recognition using audio style transfer. In R. S. Romaniuk, & M. Linczuk (Eds.), Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2019 [111762J] (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11176). SPIE. https://doi.org/10.1117/12.2536535