Work

BembaSpeech: A Speech Recognition Corpus for the Bemba Language

Year: 2022
Type: conference-paper
Abstract: We present a preprocessed, ready-to-use automatic speech recognition corpus, BembaSpeech, consisting over 24 hours of read speech in the Bemba language, a written but low-resourced language spoken by over 30% of the population in Zambia.To assess its usefulness for training and testing ASR systems for Bemba, we explored different approaches; supervised pre-training (training from scratch), cross-lingual transfer learning from a monolingual English pre-trained model using DeepSpeech on the portion of the dataset and fine-tuning large scale self-supervised Wav2Vec2.0based multilingual pre-trained models on the complete Bem-baSpeech corpus.From our experiments, the 1 billion XLS-R parameter model gives the best results.The model achieves a word error rate (WER) of 32.91%, results demonstrating that model capacity significantly improves performance and that multilingual pre-trained models transfers cross-lingual acoustic representation better than monolingual pre-trained English model on the Bem-baSpeech for the Bemba ASR.Lastly, results also show that the corpus can be used for building ASR systems for Bemba language.
Language: English
Open Access status: hybrid
OpenAlex ID: W3127072051
FWCI: 1.493
Cites: 44
Cited by: 28
Related to: 10