Corpus ID: 209862258

Transformer-based language modeling and decoding for conversational speech recognition

@article{Nassar2020TransformerbasedLM,
  title={Transformer-based language modeling and decoding for conversational speech recognition},
  author={Kareem Nassar},
  journal={ArXiv},
  year={2020},
  volume={abs/2001.01140}
}
  • Kareem Nassar
  • Published 2020
  • Computer Science, Engineering
  • ArXiv
  • 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. 

    Figures, Tables, and Topics from this paper

    References

    SHOWING 1-10 OF 16 REFERENCES
    Recurrent Neural Network Language Model Adaptation for Conversational Speech Recognition
    • 25
    • PDF
    Weighted finite-state transducers in speech recognition
    • 944
    • PDF
    The Kaldi Speech Recognition Toolkit
    • 4,268
    • PDF
    BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
    • 13,775
    • PDF
    Transformers: State-of-the-Art Natural Language Processing
    • 117
    • PDF
    Attention is All you Need
    • 15,021
    • PDF