mlsanityData to 14 Oct 2026

HypR: A comprehensive study for ASR hypothesis revising with a reference corpus

arXiv 2309.09838 ↗First posted 18 Sep 2023cs.CL cs.SD eess.AScited by 1

Not on the map. Why?

Abstract (arXiv)

With the development of deep learning, automatic speech recognition (ASR) has made significant progress. To further enhance the performance of ASR, revising recognition results is one of the lightweight but efficient manners. Various methods can be roughly classified into N-best reranking modeling and error correction modeling. The former aims to select the hypothesis with the lowest error rate from a set of candidates generated by ASR for a given input speech. The latter focuses on detecting recognition errors in a given hypothesis and correcting these errors to obtain an enhanced result. However, we observe that these studies are hardly comparable to each other, as they are usually evaluated on different corpora, paired with different ASR models, and even use different datasets to train the models. Accordingly, we first concentrate on providing an ASR hypothesis revising (HypR) dataset in this study. HypR contains several commonly used corpora (AISHELL-1, TED-LIUM 2, and LibriSpeech) and provides 50 recognition hypotheses for each speech utterance. The checkpoint models of ASR are also published. In addition, we implement and compare several classic and representative methods, showing the recent research progress in revising speech recognition results. We hope that the publicly available HypR dataset can become a reference benchmark for subsequent research and promote this field of research to an advanced level.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 19 of its 42 references that are in the map. First posted on arXiv 18 Sep 2023.

Builds on

19 of 42 references.

References · 19

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2018ESPnet: End-to-End Speech Processing Toolkitcited by 558
  3. 2015Listen, Attend and Spellcited by 543
  4. 2017AISHELL-1: An Open-Source Mandarin Speech Corpus and A Speech Recognition Baselinecited by 373
  5. 2019A spelling correction model for end-to-end speech recognitioncited by 63
  6. 2023Can Generative Large Language Models Perform ASR Error Correction?cited by 57
  7. 2021FastCorrect: Fast Error Correction with Edit Alignment for Automatic Speech Recognitioncited by 43
  8. 2022RescoreBERT: Discriminative Speech Recognition Rescoring with BERTcited by 34
  9. 2021FastCorrect 2: Fast Error Correction on Multiple Candidates for Automatic Speech Recognitioncited by 27
  10. 2021Innovative Bert-based Reranking Language Models for Speech Recognitioncited by 23
  11. 2021BART based semantic correction for Mandarin automatic speech recognition systemcited by 20
  12. 2022Error Correction in ASR using Sequence-to-Sequence Modelscited by 19
7 more
  1. 2021Non-autoregressive Transformer-based End-to-end ASR using BERTcited by 13
  2. 2022A context-aware knowledge transferring strategy for CTC-based ASRcited by 12
  3. 2019An Empirical Study of Efficient ASR Rescoring with Transformerscited by 11
  4. 2022SoftCorrect: Error Correction with Soft Detection for Automatic Speech Recognitioncited by 9
  5. 2022Transcormer: Transformer for Sentence Scoring with Sliding Language Modelingcited by 8
  6. 2020Fast and Accurate Deep Bidirectional Language Representations for Unsupervised Learningcited by 6
  7. 2023Masked Audio Text Encoders are Effective Multi-Modal Rescorerscited by 2

Built on by

Cited by year

Citing papers per year: 2024: 1.

1 paper.

  1. 2024Robust ASR Error Correction with Conservative Data Filteringcited by 3