Lanfrica Talks

Opeyemi Osakuade: To what extent do speech models capture the tones of tonal languages?

Lanfrica Talks Season 1 Episode 29

This podcast follows from the presentation at https://youtu.be/rV5bUikOe88. In this episode, we're joined by Opeyemi Osakuade, a PhD student at the University of Edinburgh, engaged in the study of speech models, with a particular focus on low-resourced languages. Opeyemi Lukman discusses the challenges of applying advanced speech representation models, like HuBERT, to tonal languages, using Yorùbá as a case study. The episode explores the strengths and limitations of HuBERT in capturing the nuances of Yorùbá's tonal system.


Follow Opeyemi Osakuade

Twitter: https://twitter.com/o_feranmi_

Linkedln: https://www.linkedin.com/in/opeyemi-osakuade/


--- Send in a voice message: https://podcasters.spotify.com/pod/show/lanfrica-talks-podcast/message