Contextualizing ASR Lattice Rescoring with Hybrid Pointer Network Language Model

@inproceedings{Liu2020ContextualizingAL,
  title={Contextualizing ASR Lattice Rescoring with Hybrid Pointer Network Language Model},
  author={Darong Liu and Chunxi Liu and F. Zhang and Gabriel Synnaeve and Yatharth Saraf and G. Zweig},
  booktitle={INTERSPEECH},
  year={2020}
}
Videos uploaded on social media are often accompanied with textual descriptions. In building automatic speech recognition (ASR) systems for videos, we can exploit the contextual information provided by such video metadata. In this paper, we explore ASR lattice rescoring by selectively attending to the video descriptions. We first use an attention based method to extract contextual vector representations of video metadata, and use these representations as part of the inputs to a neural language… Expand
4 Citations
Contextual RNN-T For Open Domain ASR
  • 5
  • PDF
Context-aware RNNLM Rescoring for Conversational Speech Recognition
  • PDF
Benchmarking LF-MMI, CTC And RNN-T Criteria For Streaming ASR
  • 3
  • PDF
Dual Application of Speech Enhancement for Automatic Speech Recognition
  • 2
  • PDF

References

SHOWING 1-10 OF 24 REFERENCES
Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond
  • 1,175
  • PDF
Pointer Sentinel Mixture Models
  • 696
  • PDF
Get To The Point: Summarization with Pointer-Generator Networks
  • 1,717
  • Highly Influential
  • PDF
Joint Grapheme and Phoneme Embeddings for Contextual End-to-End ASR
  • 7
  • PDF
Composition-based on-the-fly rescoring for salient n-gram biasing
  • 36
  • PDF
Two Efficient Lattice Rescoring Methods Using Recurrent Neural Network Language Models
  • 41
  • PDF
Deep Context: End-to-end Contextual Speech Recognition
  • 35
  • Highly Influential
  • PDF
Language Modeling with Deep Transformers
  • 59
  • PDF
A Pruned Rnnlm Lattice-Rescoring Algorithm for Automatic Speech Recognition
  • 61
  • Highly Influential
  • PDF
Effective Approaches to Attention-based Neural Machine Translation
  • 4,766
  • PDF
...
1
2
3
...