NWU Institutional Repository

Collecting and evaluating speech recognition corpora for 11 South African languages

Loading...
Thumbnail Image

Date

Authors

Badenhorst, Jaco
Van Heerden, Charl
Barnard, Etienne
Davel, Marelie H.

Journal Title

Journal ISSN

Volume Title

Publisher

Springer

Abstract

We describe the Lwazi corpus for automatic speech recognition (ASR), a new telephone speech corpus which contains data from the eleven official languages of South Africa. Because of practical constraints, the amount of speech per language is relatively small compared to major corpora in world languages, and we report on our investigation of the stability of the ASR models derived from the corpus. We also report on phoneme distance measures across languages, and describe initial phone recognisers that were developed using this data. We find that a surprisingly small number of speakers (fewer than 50) and around 10 to 20 h of speech per language are sufficient for the purposes of acceptable phone-based recognition.

Description

Citation

Badenhorst, J. & Van Heerden, C., et al. 2011. Collecting and evaluating speech recognition corpora for 11 South African languages. Language resources and evaluation, 45(3):289-309. [http://link.springer.com/journal/10579]

Endorsement

Review

Supplemented By

Referenced By