Improving Domain-Specific ASR with LLM-Generated Contextual Descriptions
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_ | 1866913445209702400 |
|---|---|
| author | Suh, Jiwon Na, Injae Jung, Woohwan |
| author_facet | Suh, Jiwon Na, Injae Jung, Woohwan |
| contents | End-to-end automatic speech recognition (E2E ASR) systems have significantly improved speech recognition through training on extensive datasets. Despite these advancements, they still struggle to accurately recognize domain specific words, such as proper nouns and technical terminologies. To address this problem, we propose a method to utilize the state-of-the-art Whisper without modifying its architecture, preserving its generalization performance while enabling it to leverage descriptions effectively. Moreover, we propose two additional training techniques to improve the domain specific ASR: decoder fine-tuning, and context perturbation. We also propose a method to use a Large Language Model (LLM) to generate descriptions with simple metadata, when descriptions are unavailable. Our experiments demonstrate that proposed methods notably enhance domain-specific ASR accuracy on real-life datasets, with LLM-generated descriptions outperforming human-crafted ones in effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_17874 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Improving Domain-Specific ASR with LLM-Generated Contextual Descriptions Suh, Jiwon Na, Injae Jung, Woohwan Computation and Language Artificial Intelligence End-to-end automatic speech recognition (E2E ASR) systems have significantly improved speech recognition through training on extensive datasets. Despite these advancements, they still struggle to accurately recognize domain specific words, such as proper nouns and technical terminologies. To address this problem, we propose a method to utilize the state-of-the-art Whisper without modifying its architecture, preserving its generalization performance while enabling it to leverage descriptions effectively. Moreover, we propose two additional training techniques to improve the domain specific ASR: decoder fine-tuning, and context perturbation. We also propose a method to use a Large Language Model (LLM) to generate descriptions with simple metadata, when descriptions are unavailable. Our experiments demonstrate that proposed methods notably enhance domain-specific ASR accuracy on real-life datasets, with LLM-generated descriptions outperforming human-crafted ones in effectiveness. |
| title | Improving Domain-Specific ASR with LLM-Generated Contextual Descriptions |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2407.17874 |