mlsanityData to 14 Oct 2026

Fast Entropy-Based Methods of Word-Level Confidence Estimation for End-To-End Automatic Speech Recognition

arXiv 2212.08703 ↗First posted 16 Dec 2022eess.AS cs.CL cs.ITcited by 7

Not on the map. Why?

Abstract (arXiv)

This paper presents a class of new fast non-trainable entropy-based confidence estimation methods for automatic speech recognition. We show how per-frame entropy values can be normalized and aggregated to obtain a confidence measure per unit and per word for Connectionist Temporal Classification (CTC) and Recurrent Neural Network Transducer (RNN-T) models. Proposed methods have similar computational complexity to the traditional method based on the maximum per-frame probability, but they are more adjustable, have a wider effective threshold range, and better push apart the confidence distributions of correct and incorrect words. We evaluate the proposed confidence measures on LibriSpeech test sets, and show that they are up to 2 and 4 times better than confidence estimation based on the maximum per-frame probability at detecting incorrect words for Conformer-CTC and Conformer-RNN-T models, respectively.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 17 of its 41 references that are in the map. First posted on arXiv 16 Dec 2022.

Builds on

17 of 41 references.

References · 17

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  2. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
  3. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
  4. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 776
  5. 2015MUSAN: A Music, Speech, and Noise Corpuscited by 425
  6. 2019NeMo: a toolkit for building AI applications using Neural Modulescited by 235
  7. 2020Improved Noisy Student Training for Automatic Speech Recognitioncited by 141
  8. 2021SleepTransformer: Automatic Sleep Staging with Interpretability and Uncertainty Quantificationcited by 84
  9. 2020Confidence Estimation for Attention-based Sequence-to-sequence Models for Speech Recognitioncited by 35
  10. 2021Learning Word-Level Confidence For Subword End-to-End ASRcited by 22
  11. 2021An evaluation of word-level confidence estimation for end-to-end automatic speech recognitioncited by 18
  12. 2019Active and Semi-Supervised Learning in ASR: Benefits on the Acoustic and Language Modelscited by 16
5 more
  1. 2021Residual Energy-Based Models for End-to-End Speech Recognitioncited by 9
  2. 2021Multi-Task Learning for End-to-End ASR Word and Utterance Confidence with Deletion Predictioncited by 8
  3. 2019On Modeling ASR Word Confidencecited by 6
  4. 2021Entropy methods for the confidence assessment of probabilistic classification modelscited by 4
  5. 2021Word-level confidence estimation for RNN transducerscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 3; 2025: 2; 2026: 1.

7 papers.

  1. 2024Empowering Low-Resource Language ASR via Large-Scale Pseudo Labelingcited by 5
  2. 2023Confidence-based Ensembles of End-to-End Speech Recognition Modelscited by 4
  3. 2024BaFTA: Backprop-Free Test-Time Adaptation For Zero-Shot Vision-Language Modelscited by 2
  4. 2025Adopting Whisper for Confidence Estimationcited by 2
  5. 2024TeLeS: Temporal Lexeme Similarity Score to Estimate Confidence in End-to-End ASRcited by 1
  6. 2025Confidence-Guided Error Correction for Disordered Speech Recognitioncited by 1
  7. 2026Enhancing ASR Performance in the Medical Domain for Dravidian Languagescited by 0