Corpus ID: 207863393

Long-span language modeling for speech recognition

@article{Parthasarathy2019LongspanLM,
  title={Long-span language modeling for speech recognition},
  author={S. Parthasarathy and W. Gale and X. Chen and George Polovets and S. Chang},
  journal={ArXiv},
  year={2019},
  volume={abs/1911.04571}
}
We explore neural language modeling for speech recognition where the context spans multiple sentences. Rather than encode history beyond the current sentence using a cache of words or document-level features, we focus our study on the ability of LSTM and Transformer language models to implicitly learn to carry over context across sentence boundaries. We introduce a new architecture that incorporates an attention mechanism into LSTM to combine the benefits of recurrent and attention… Expand
3 Citations

References

SHOWING 1-10 OF 28 REFERENCES
Session-level Language Modeling for Conversational Speech
  • 10
  • PDF
Language Modeling with Deep Transformers
  • 55
  • Highly Influential
  • PDF
LSTM Neural Networks for Language Modeling
  • 1,245
  • PDF
Context dependent recurrent neural network language model
  • 488
  • PDF
Cache based recurrent neural network language model inference for first pass speech recognition
  • 40
  • PDF
Exploring the Limits of Language Modeling
  • 804
  • PDF
A Cache-Based Natural Language Model for Speech Recognition
  • R. Kuhn, R. Mori
  • Computer Science
  • IEEE Trans. Pattern Anal. Mach. Intell.
  • 1990
  • 569
Hierarchical Recurrent Neural Network for Document Modeling
  • 129
  • PDF
Effectively Building Tera Scale MaxEnt Language Models Incorporating Non-Linguistic Signals
  • 19
  • PDF
Efficient lattice rescoring using recurrent neural network language models
  • 79
  • PDF
...
1
2
3
...