mlsanityData to 7 Oct 2026

Large vocabulary speech recognition for languages of Africa: multilingual modeling and self-supervised learning

arXiv 2208.03067 ↗First posted 5 Aug 2022cs.CL cs.SD eess.AScited by 5

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 17 of its 44 references that are in the map. First posted on arXiv 5 Aug 2022.

Builds on

17 of 44 references.

References · 17

  1. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,853
  2. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,838
  3. 2019Common Voice: A Massively-Multilingual Speech Corpuscited by 1,358
  4. 2018Streaming End-to-end Speech Recognition For Mobile Devicescited by 230
  5. 2020Hybrid Autoregressive Transducer (hat)cited by 106
  6. 2021BigSSL: Exploring the Frontier of Large-Scale Semi-Supervised Learning for Automatic Speech Recognitioncited by 88
  7. 2017Multilingual Speech Recognition With A Single End-To-End Modelcited by 87
  8. 2020FastEmit: Low-latency Streaming ASR with Sequence-level Emission Regularizationcited by 66
  9. 2021Scaling End-to-End Models for Large-Scale Multilingual ASRcited by 61
  10. 2018Multilingual End-to-End Speech Recognition with A Single Transformer on Low-Resource Languagescited by 26
  11. 2021Large-scale ASR Domain Adaptation using Self- and Semi-supervised Learningcited by 26
  12. 2018Sequence-based Multi-lingual Low Resource Speech Recognitioncited by 20
5 more
  1. 2022Accented Speech Recognition: Benchmarking, Pre-training, and Diverse Datacited by 15
  2. 2021BembaSpeech: A Speech Recognition Corpus for the Bemba Languagecited by 14
  3. 2021OkwuGb'e: End-to-End Speech Recognition for Fon and Igbocited by 13
  4. 2019Improved low-resource Somali speech recognition by semi-supervised acoustic and language model trainingcited by 3
  5. 2021Text Normalization for Low-Resource Languages of Africacited by 2

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 1; 2025: 1.

5 papers.

  1. 2023MasakhaPOS: Part-of-Speech Tagging for Typologically Diverse African Languagescited by 23
  2. 2024Africa-Centric Self-Supervised Pre-Training for Multilingual Speech Representation in a Sub-Saharan Contextcited by 8
  3. 2023AfriNames: Most ASR models "butcher" African Namescited by 7
  4. 2023KinSPEAK: Improving speech recognition for Kinyarwanda via semi-supervised learning methodscited by 2
  5. 2025Benchmarking Automatic Speech Recognition Models for African Languagescited by 1