FastCorrect 2: Fast Error Correction on Multiple Candidates for Automatic Speech Recognition

Leng, Yichong; Tan, Xu; Wang, Rui; Zhu, Linchen; Xu, Jinan; Liu, Linquan; Qin, Tao; Li, Xiangyang; Lin, Edward; Liu, Tie-Yan

doi:10.48550/arxiv.2109.14420

Cited by 2 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Team affecting the effect of simultaneous translation in real scenes. Some work has attempted to improve the pipeline systems by introducing an ASR error correction model (Leng et al, 2021;Zhang et al, 2021a), others proposed pre-training approaches to alleviate the data scarcity problem of speech translation corpora in end-to-end systems Pino et al, 2020;Zheng et al, 2021;Li et al, 2020b;. We hope to see more participants in future workshops investigating how to close the performance gap between the two tracks.…”

Section: Rankmentioning

confidence: 97%

Findings of the Second Workshop on Automatic Simultaneous Translation

Zhang¹,

Zhang²,

He³

et al. 2021

Proceedings of the Second Workshop on Automatic Simultaneous Translation

View full text Add to dashboard Cite

This paper presents the results of the shared task of the 2nd Workshop on Automatic Simultaneous Translation (AutoSimTrans). The task includes two tracks, one for text-to-text translation and one for speech-to-text, requiring participants to build systems to translate from either the source text or speech into the target text. Different from traditional machine translation, the AutoSimTrans shared task evaluates not only translation quality but also latency. We propose a metric "Monotonic Optimal Sequence" (MOS) considering both quality and latency to rank the submissions. We also discuss some important open issues in simultaneous translation.

show abstract

Section: Rankmentioning

confidence: 97%

Findings of the Second Workshop on Automatic Simultaneous Translation

Zhang¹,

Zhang²,

He³

et al. 2021

Proceedings of the Second Workshop on Automatic Simultaneous Translation

View full text Add to dashboard Cite

show abstract

“…To this end, researchers suggested text-to-text error correction methods specifically designed to correct transcribed errors (Mani et al, 2020;Leng et al, 2021a;Hrinchuk et al, 2020). These transformer-based methods typically requires a massive and diverse training data (i.e., transcribed audio segments and their corresponding humanauthored transcriptions) which is rarely at reach.…”

Section: Introductionmentioning

confidence: 99%

Optimized Tokenization for Transcribed Error Correction

Wullach,

Chazan

2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

The challenges facing speech recognition systems, such as variations in pronunciations, adverse audio conditions, and the scarcity of labeled data, emphasize the necessity for a postprocessing step that corrects recurring errors. Previous research has shown the advantages of employing dedicated error correction models, yet training such models requires large amounts of labeled data which is not easily obtained. To overcome this limitation, synthetic transcribedlike data is often utilized, however, bridging the distribution gap between transcribed errors and synthetic noise is not trivial. In this paper, we demonstrate that the performance of correction models can be significantly increased by training solely using synthetic data. Specifically, we empirically show that: (1) synthetic data generated using the error distribution derived from a set of transcribed data outperforms the common approach of applying random perturbations; (2) applying language-specific adjustments to the vocabulary of a BPE tokenizer strike a balance between adapting to unseen distributions and retaining knowledge of transcribed errors. We showcase the benefits of these key observations, and evaluate our approach using multiple languages, speech recognition systems and prominent speech recognition datasets.

show abstract

FastCorrect 2: Fast Error Correction on Multiple Candidates for Automatic Speech Recognition

Abstract: formance than the cascaded re-scoring and correction pipeline and can serve as a unified postprocessing module for ASR.

Cited by 2 publications

References 23 publications

Findings of the Second Workshop on Automatic Simultaneous Translation

Findings of the Second Workshop on Automatic Simultaneous Translation

Optimized Tokenization for Transcribed Error Correction

Contact Info

Product

Resources

About