Pinyin Regularization in Error Correction for Chinese Speech Recognition with Large Language Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914955559698432 |
|---|---|
| author | Tang, Zhiyuan Wang, Dong Huang, Shen Shang, Shidong |
| author_facet | Tang, Zhiyuan Wang, Dong Huang, Shen Shang, Shidong |
| contents | Recent studies have demonstrated the efficacy of large language models (LLMs) in error correction for automatic speech recognition (ASR). However, much of the research focuses on the English language. This paper redirects the attention to Chinese. Firstly, we construct a specialized benchmark dataset aimed at error correction for Chinese ASR with 724K hypotheses-transcription pairs, named the Chinese Hypotheses Paradise dataset (ChineseHP), which contains a wide range of scenarios and presents significant challenges. Subsequently, we conduct a preliminary evaluation using the dataset for both direct-prompting and fine-tuning pre-trained LLMs. Furthermore, we propose a straightforward method of Pinyin regularization for prompts, which involves the transcription of Pinyin directly from text hypotheses. The experimental results reveal that Pinyin regularization consistently enhances the error-correcting ability of LLMs when compared with those without regularization. The dataset is available on the website. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_01909 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Pinyin Regularization in Error Correction for Chinese Speech Recognition with Large Language Models Tang, Zhiyuan Wang, Dong Huang, Shen Shang, Shidong Computation and Language Sound Audio and Speech Processing Recent studies have demonstrated the efficacy of large language models (LLMs) in error correction for automatic speech recognition (ASR). However, much of the research focuses on the English language. This paper redirects the attention to Chinese. Firstly, we construct a specialized benchmark dataset aimed at error correction for Chinese ASR with 724K hypotheses-transcription pairs, named the Chinese Hypotheses Paradise dataset (ChineseHP), which contains a wide range of scenarios and presents significant challenges. Subsequently, we conduct a preliminary evaluation using the dataset for both direct-prompting and fine-tuning pre-trained LLMs. Furthermore, we propose a straightforward method of Pinyin regularization for prompts, which involves the transcription of Pinyin directly from text hypotheses. The experimental results reveal that Pinyin regularization consistently enhances the error-correcting ability of LLMs when compared with those without regularization. The dataset is available on the website. |
| title | Pinyin Regularization in Error Correction for Chinese Speech Recognition with Large Language Models |
| topic | Computation and Language Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2407.01909 |