Third Workshop on Spoken Language Technologies for Under-resourced Languages Cape Town, South Africa May 7-9, 2012

Tone Realisation in a Yorùbá Speech Recognition Corpus

Daniel R. van Niekerk (1,2), Etienne Barnard (2)

(1) Human Language Technologies Research Group, Meraka Institute, CSIR, Pretoria, South Africa.
(2) Multilingual Speech Technologies, North-West University, Vanderbijlpark, South Africa.

We investigate the acoustic realisation of tone in short continuous utterances in Yorùbá. Fundamental frequency (F0) contours are extracted for automatically aligned syllables from a speech corpus of 33 speakers collected for speech recognition development. Extracted contours are processed and analysed statistically to describe acoustic properties in different tonal contexts. We demonstrate how features useful for tone recognition or synthesis can be successfully extracted from a corpus of this nature and confirm some previously described phenomena in this context.

Index Terms: Yorùbá, tone language, fundamental frequency

Full Paper

Bibliographic reference.  Niekerk, Daniel R. van / Barnard, Etienne (2012): "Tone realisation in a yorùbá speech recognition corpus", In SLTU-2012, 54-59.