Transcribing and Annotating Speech Corpora for Speech Recognition: A Three-Step Crowdsourcing Approach with Quality Control

Authors

  • Annika Hämäläinen Microsoft Language Development Center
  • Fernando Pinto Moreira Microsoft Language Development Center
  • Jairo Avelar Microsoft Language Development Center
  • Daniela Braga Microsoft Language Development Center
  • Miguel Sales Dias Microsoft Language Development Center

DOI:

https://doi.org/10.1609/hcomp.v1i1.13102

Keywords:

automatic speech recognition, speech corpora, transcription, annotation, crowdsourcing

Abstract

Large speech corpora with word-level transcriptions annotated for noises and disfluent speech are necessary for training automatic speech recognisers. Crowdsourcing is a lower-cost, faster-turnaround, highly scalable alternative for expert transcription and annotation. In this paper, we showcase our three-step crowdsourcing approach motivated by the importance of accurate transcriptions and annotations.

Downloads

Published

2013-11-03

How to Cite

Hämäläinen, A., Pinto Moreira, F., Avelar, J., Braga, D., & Sales Dias, M. (2013). Transcribing and Annotating Speech Corpora for Speech Recognition: A Three-Step Crowdsourcing Approach with Quality Control. Proceedings of the AAAI Conference on Human Computation and Crowdsourcing, 1(1), 30-31. https://doi.org/10.1609/hcomp.v1i1.13102