OpenASR20: An Open Challenge for Automatic Speech Recognition ofConversational Telephone Speech in Low-Resource Languages

Published

September 1, 2021

Author(s)

Kay Peterson, Audrey N. Tong, Jennifer Yu

Abstract

In 2020, the National Institute of Standards and Technology (NIST), in cooperation with the Intelligence Advanced Research Project Activity (IARPA), conducted an open challenge on automatic speech recognition (ASR) technology for low-resource languages on a challenging data type - conversational telephone speech. The OpenASR20 Challenge was offered for ten low-resource languages - Amharic, Cantonese, Guarani, Javanese, Kurmanji Kurdish, Mongolian, Pashto, Somali, Tamil, and Vietnamese. A total of nine teams from five countries fully participated, and 128 valid submissions were scored. This paper gives an overview of the challenge setup and procedures, as well as a summary of the results. The results show overall high word error rate (WER), with the best results on a severely constrained training data condition ranging from 0.4 to 0.65, depending on the language. ASR with such limited resources remains a challenging problem. Providing a computing platform may be a way to level the playing field and encourage wider participation in challenges like OpenASR.

Proceedings Title

Proc. Interspeech 2021

Conference Dates

August 31-September 3, 2021

Conference Location

Brno, CZ

Conference Title

INTERSPEECH 2021, Special session on OpenASR and Low Resource ASR Development

Pub Type

Conferences

Download Paper

https://doi.org/10.21437/Interspeech.2021-1930

Local Download

Keywords

automatic speech recognition, evaluation, low-resource language, conversational telephone speech, IARPA MATERIAL, Amharic, Cantonese, Guarani, Javanese, Kurmanji Kurdish, Mongolian, Pashto, Somali, Tamil, Vietnamese

Citation

Peterson, K. , Tong, A. and Yu, J. (2021), OpenASR20: An Open Challenge for Automatic Speech Recognition ofConversational Telephone Speech in Low-Resource Languages, Proc. Interspeech 2021, Brno, CZ, [online], https://doi.org/10.21437/Interspeech.2021-1930, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=932302 (Accessed December 17, 2024)