English

Transformer-based language modeling and decoding for conversational speech recognition

Computation and Language 2020-01-07 v1 Machine Learning Audio and Speech Processing

Abstract

We propose a way to use a transformer-based language model in conversational speech recognition. Specifically, we focus on decoding efficiently in a weighted finite-state transducer framework. We showcase an approach to lattice re-scoring that allows for longer range history captured by a transfomer-based language model and takes advantage of a transformer's ability to avoid computing sequentially.

Keywords

Cite

@article{arxiv.2001.01140,
  title  = {Transformer-based language modeling and decoding for conversational speech recognition},
  author = {Kareem Nassar},
  journal= {arXiv preprint arXiv:2001.01140},
  year   = {2020}
}
R2 v1 2026-06-23T13:02:57.717Z