Large vocabulary speech recognition for languages of Africa: multilingual modeling and self-supervised learning
This addresses the problem of limited speech recognition access for speakers of many African languages, but it is incremental as it builds on existing techniques.
The paper tackled the lack of automatic speech recognition systems for African languages by experimenting with multilingual modeling and self-supervised learning on data from 15 languages, showing that these techniques improve speech recognition quality.
Almost none of the 2,000+ languages spoken in Africa have widely available automatic speech recognition systems, and the required data is also only available for a few languages. We have experimented with two techniques which may provide pathways to large vocabulary speech recognition for African languages: multilingual modeling and self-supervised learning. We gathered available open source data and collected data for 15 languages, and trained experimental models using these techniques. Our results show that pooling the small amounts of data available in multilingual end-to-end models, and pre-training on unsupervised data can help improve speech recognition quality for many African languages.