Corpus ID: 232185120

Learning Word-Level Confidence For Subword End-to-End ASR

@article{Qiu2021LearningWC,
  title={Learning Word-Level Confidence For Subword End-to-End ASR},
  author={David Qiu and Qiujia Li and Yanzhang He and Y. Zhang and Bo Li and L. Cao and Rohit Prabhavalkar and Deepti Bhatia and Wei Li and Ke Hu and T. Sainath and Ian McGraw},
  journal={ArXiv},
  year={2021},
  volume={abs/2103.06716}
}
We study the problem of word-level confidence estimation in subword-based end-to-end (E2E) models for automatic speech recognition (ASR). Although prior works have proposed training auxiliary confidence models for ASR systems, they do not extend naturally to systems that operate on word-pieces (WP) as their vocabulary. In particular, ground truth WP correctness labels are needed for training confidence models, but the non-unique tokenization from word to WP causes inaccurate labels to be… Expand
4 Citations

Figures and Tables from this paper

Multi-Task Learning for End-to-End ASR Word and Utterance Confidence with Deletion Prediction
  • PDF
Residual Energy-Based Models for End-to-End Speech Recognition
  • 1
  • PDF
On Addressing Practical Challenges for RNN-Transducer
  • PDF
Understanding Medical Conversations: Rich Transcription, Confidence Scores & Information Extraction
  • PDF

References

SHOWING 1-10 OF 40 REFERENCES
Confidence Estimation for Attention-based Sequence-to-sequence Models for Speech Recognition
  • 3
  • PDF
Bi-directional Lattice Recurrent Neural Networks for Confidence Estimation
  • Q. Li, Preben Ness, A. Ragni, M. Gales
  • Computer Science, Engineering
  • ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2019
  • 14
  • PDF
Deliberation Model Based Two-Pass End-To-End Speech Recognition
  • 15
  • PDF
Finding consensus in speech recognition: word error minimization and other applications of confusion networks
  • 716
  • PDF
Towards Better Decoding and Language Model Integration in Sequence to Sequence Models
  • 207
  • PDF
Confidence measures for large vocabulary continuous speech recognition
  • 492
  • PDF
A Streaming On-Device End-To-End Model Surpassing Server-Side Conventional Model Quality and Latency
  • T. Sainath, Yanzhang He, +26 authors D. Zhao
  • Computer Science
  • ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2020
  • 65
  • PDF
Calibration of Confidence Measures in Speech Recognition
  • Dong Yu, J. Li, Li Deng
  • Mathematics
  • IEEE Transactions on Audio, Speech, and Language Processing
  • 2011
  • 56
  • PDF
BPE-Dropout: Simple and Effective Subword Regularization
  • 44
  • PDF
...
1
2
3
4
...