Language-Aware Prompt Tuning for Parameter-Efficient Seamless Language Expansion in Multilingual ASR

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Yang, Hongli, Li, Sheng, Huang, Hao, Tuohan, Ayiduosi, Peng, Yizhou
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915515757232128
author Yang, Hongli
Li, Sheng
Huang, Hao
Tuohan, Ayiduosi
Peng, Yizhou
author_facet Yang, Hongli
Li, Sheng
Huang, Hao
Tuohan, Ayiduosi
Peng, Yizhou
contents Recent advancements in multilingual automatic speech recognition (ASR) have been driven by large-scale end-to-end models like Whisper. However, challenges such as language interference and expanding to unseen languages (language expansion) without degrading performance persist. This paper addresses these with three contributions: 1) Entire Soft Prompt Tuning (Entire SPT), which applies soft prompts to both the encoder and decoder, enhancing feature extraction and decoding; 2) Language-Aware Prompt Tuning (LAPT), which leverages cross-lingual similarities to encode shared and language-specific features using lightweight prompt matrices; 3) SPT-Whisper, a toolkit that integrates SPT into Whisper and enables efficient continual learning. Experiments across three languages from FLEURS demonstrate that Entire SPT and LAPT outperform Decoder SPT by 5.0% and 16.0% in language expansion tasks, respectively, providing an efficient solution for dynamic, multilingual ASR models with minimal computational overhead.
format Preprint
id arxiv_https___arxiv_org_abs_2506_21577
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Language-Aware Prompt Tuning for Parameter-Efficient Seamless Language Expansion in Multilingual ASR
Yang, Hongli
Li, Sheng
Huang, Hao
Tuohan, Ayiduosi
Peng, Yizhou
Computation and Language
Artificial Intelligence
Sound
Audio and Speech Processing
Recent advancements in multilingual automatic speech recognition (ASR) have been driven by large-scale end-to-end models like Whisper. However, challenges such as language interference and expanding to unseen languages (language expansion) without degrading performance persist. This paper addresses these with three contributions: 1) Entire Soft Prompt Tuning (Entire SPT), which applies soft prompts to both the encoder and decoder, enhancing feature extraction and decoding; 2) Language-Aware Prompt Tuning (LAPT), which leverages cross-lingual similarities to encode shared and language-specific features using lightweight prompt matrices; 3) SPT-Whisper, a toolkit that integrates SPT into Whisper and enables efficient continual learning. Experiments across three languages from FLEURS demonstrate that Entire SPT and LAPT outperform Decoder SPT by 5.0% and 16.0% in language expansion tasks, respectively, providing an efficient solution for dynamic, multilingual ASR models with minimal computational overhead.
title Language-Aware Prompt Tuning for Parameter-Efficient Seamless Language Expansion in Multilingual ASR
topic Computation and Language
Artificial Intelligence
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2506.21577