/Pseudo-Labeling-for-Domain-Agnostic-Bangla-ASR

Pseudo-Labeling for Domain-Agnostic Bangla Automatic Speech Recognition

OtherNOASSERTION

Pseudo-Labeling-for-Domain-Agnostic-Bangla-ASR

Pseudo-Labeling for Domain-Agnostic Bangla Automatic Speech Recognition

Pretraiend model

Trained on 18k+ hours psedu-annotated Bangla ASR Corpus

https://huggingface.co/hishab/hishab_bn_fastconformer

Test data

https://huggingface.co/datasets/hishab/MegaBNSpeech_Test_Data

Pseudo-Labeling for Domain-Agnostic Bangla Automatic Speech Recognition

Authors: Rabindra Nath Nandi, Mehadi Hasan Menon, Tareq Al Muntasir, Sagor Sarker, Quazi Sarwar Muhtaseem, Md. Tariqul Islam, Shammur Absar Chowdhury, Firoj Alam

Affiliations:

  • Hishab Singapore Pte. Ltd, Singapore
  • Qatar Computing Research Institute, HBKU, Doha, Qatar

Accepted in: The First Bangla Language Processing Workshop (BLP 2023) Co-located with EMNLP 2023 in Singapore

Citation

If you find this work useful in your research, please consider citing the paper: