Speak & Spell: LLM-Driven Controllable Phonetic Error Augmentation for Robust Dialogue State Tracking
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909876851048448 |
|---|---|
| author | Lee, Jihyun Im, Solee Lee, Wonjun Lee, Gary Geunbae |
| author_facet | Lee, Jihyun Im, Solee Lee, Wonjun Lee, Gary Geunbae |
| contents | Dialogue State Tracking (DST) is a key part of task-oriented dialogue systems, identifying important information in conversations. However, its accuracy drops significantly in spoken dialogue environments due to named entity errors from Automatic Speech Recognition (ASR) systems. We introduce a simple yet effective data augmentation method that targets those entities to improve the robustness of DST model. Our novel method can control the placement of errors using keyword-highlighted prompts while introducing phonetically similar errors. As a result, our method generated sufficient error patterns on keywords, leading to improved accuracy in noised and low-accuracy ASR environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_06263 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Speak & Spell: LLM-Driven Controllable Phonetic Error Augmentation for Robust Dialogue State Tracking Lee, Jihyun Im, Solee Lee, Wonjun Lee, Gary Geunbae Computation and Language Artificial Intelligence Dialogue State Tracking (DST) is a key part of task-oriented dialogue systems, identifying important information in conversations. However, its accuracy drops significantly in spoken dialogue environments due to named entity errors from Automatic Speech Recognition (ASR) systems. We introduce a simple yet effective data augmentation method that targets those entities to improve the robustness of DST model. Our novel method can control the placement of errors using keyword-highlighted prompts while introducing phonetically similar errors. As a result, our method generated sufficient error patterns on keywords, leading to improved accuracy in noised and low-accuracy ASR environments. |
| title | Speak & Spell: LLM-Driven Controllable Phonetic Error Augmentation for Robust Dialogue State Tracking |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2409.06263 |