BembaSpeech: A Speech Recognition Corpus for the Bemba Language
This corpus addresses the lack of speech recognition resources for the Bemba language, a low-resourced language spoken by a significant population in Zambia.
This paper introduces BembaSpeech, a new speech recognition corpus for the Bemba language, comprising over 24 hours of read speech. They fine-tuned a pre-trained DeepSpeech English model on this corpus, achieving a Word Error Rate (WER) of 54.78%.
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 train an end-to-end Bemba ASR system by fine-tuning a pre-trained DeepSpeech English model on the training portion of the BembaSpeech corpus. Our best model achieves a word error rate (WER) of 54.78%. The results show that the corpus can be used for building ASR systems for Bemba. The corpus and models are publicly released at https://github.com/csikasote/BembaSpeech.