ISCA Archive IberSPEECH 2022
ISCA Archive IberSPEECH 2022

BCN2BRNO: ASR System Fusion for Albayzin 2022 Speech to Text Challenge

Martin Kocour, Jahnavi Umesh, Martin Karafiat, Ján Švec, Fernando López, Jordi Luque, Karel Beneš, Mireia Diez, Igor Szoke, Karel Veselý, Lukáš Burget, Jan Černocký

This paper describes the joint effort of BUT and Telefónica Research on the development of Automatic Speech Recognition systems for the Albayzin 2022 Challenge. We train and evaluate both hybrid systems and those based on end-to-end models. We also investigate the use of self-supervised learning speech representations from pre-trained models and their impact on ASR performance (as opposed to training models directly from scratch). Additionally, we also apply the Whisper model in a zero-shot fashion, postprocessing its output to fit the required transcription format. On top of tuning the model architectures and overall training schemes, we improve the robustness of our models by augmenting the training data with noises extracted from the target domain. Moreover, we apply rescoring with an external LM on top of N-best hypotheses to adjust each sentence score and pick the single best hypothesis. All these efforts lead to a significant WER reduction. Our single best system and the fusion of selected systems achieved 16.3% and 13.7% WER respectively on RTVE2020 test partition, i.e. the official evaluation partition from the previous Albayzin challenge.


doi: 10.21437/IberSPEECH.2022-56

Cite as: Kocour, M., Umesh, J., Karafiat, M., Švec, J., López, F., Luque, J., Beneš, K., Diez, M., Szoke, I., Veselý, K., Burget, L., Černocký, J. (2022) BCN2BRNO: ASR System Fusion for Albayzin 2022 Speech to Text Challenge. Proc. IberSPEECH 2022, 276-280, doi: 10.21437/IberSPEECH.2022-56

@inproceedings{kocour22_iberspeech,
  author={Martin Kocour and Jahnavi Umesh and Martin Karafiat and Ján Švec and Fernando López and Jordi Luque and Karel Beneš and Mireia Diez and Igor Szoke and Karel Veselý and Lukáš Burget and Jan Černocký},
  title={{BCN2BRNO: ASR System Fusion for Albayzin 2022 Speech to Text Challenge}},
  year=2022,
  booktitle={Proc. IberSPEECH 2022},
  pages={276--280},
  doi={10.21437/IberSPEECH.2022-56}
}