@inproceedings{lee-etal-2025-hakka,
title = "{H}akka Speech Recognition with Whisper and {P}inyin Post-processing for {FSR}-2025",
author = "Lee, Chia-Hsin and
Chang, Yung-Jun and
Wu, Jin-Yan and
Chen, Kuan-Yu",
editor = "Chang, Kai-Wei and
Lu, Ke-Han and
Yang, Chih-Kai and
Tam, Zhi-Rui and
Chang, Wen-Yu and
Wang, Chung-Che",
booktitle = "Proceedings of the 37th Conference on Computational Linguistics and Speech Processing (ROCLING 2025)",
month = nov,
year = "2025",
address = "National Taiwan University, Taipei City, Taiwan",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.rocling-main.53/",
pages = "454--458",
ISBN = "979-8-89176-379-1",
abstract = "本研究為參加 FSR-2025 客語語音辨識挑戰賽(Hakka ASR II)的技術報告,旨在推進客語自動語音辨識技術的發展。由於客語屬於低資源語言,且存在多種腔調,語音辨識面臨高度挑戰。我們以 Whisperlarge-v2 為骨幹模型,設計兩階段訓練流程:首先利用「Hakka Across Taiwan(HAT)」語料庫進行模型調適,以捕捉客語的一般聲學特徵;其次在賽事方提供的60 小時腔調語料上進行微調,以增強對目標資料的適應性。實驗發現,直接輸出客語漢字可達到良好的字錯率(CER),但由 於腔調差異與拼音規則變化多,拼音任務表現顯著下降。為解決此問題,我們以漢字模型的編碼器初始化拼音模型,並提出結合 RoBERTa 漢字轉拼音、腔調判斷與字典修正的後處理模組,期望可以在比賽中提升辨識的成效。"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lee-etal-2025-hakka">
<titleInfo>
<title>Hakka Speech Recognition with Whisper and Pinyin Post-processing for FSR-2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chia-Hsin</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yung-Jun</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jin-Yan</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kuan-Yu</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 37th Conference on Computational Linguistics and Speech Processing (ROCLING 2025)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kai-Wei</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ke-Han</namePart>
<namePart type="family">Lu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chih-Kai</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhi-Rui</namePart>
<namePart type="family">Tam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wen-Yu</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chung-Che</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">National Taiwan University, Taipei City, Taiwan</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-379-1</identifier>
</relatedItem>
<abstract>本研究為參加 FSR-2025 客語語音辨識挑戰賽(Hakka ASR II)的技術報告,旨在推進客語自動語音辨識技術的發展。由於客語屬於低資源語言,且存在多種腔調,語音辨識面臨高度挑戰。我們以 Whisperlarge-v2 為骨幹模型,設計兩階段訓練流程:首先利用「Hakka Across Taiwan(HAT)」語料庫進行模型調適,以捕捉客語的一般聲學特徵;其次在賽事方提供的60 小時腔調語料上進行微調,以增強對目標資料的適應性。實驗發現,直接輸出客語漢字可達到良好的字錯率(CER),但由 於腔調差異與拼音規則變化多,拼音任務表現顯著下降。為解決此問題,我們以漢字模型的編碼器初始化拼音模型,並提出結合 RoBERTa 漢字轉拼音、腔調判斷與字典修正的後處理模組,期望可以在比賽中提升辨識的成效。</abstract>
<identifier type="citekey">lee-etal-2025-hakka</identifier>
<location>
<url>https://aclanthology.org/2025.rocling-main.53/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>454</start>
<end>458</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Hakka Speech Recognition with Whisper and Pinyin Post-processing for FSR-2025
%A Lee, Chia-Hsin
%A Chang, Yung-Jun
%A Wu, Jin-Yan
%A Chen, Kuan-Yu
%Y Chang, Kai-Wei
%Y Lu, Ke-Han
%Y Yang, Chih-Kai
%Y Tam, Zhi-Rui
%Y Chang, Wen-Yu
%Y Wang, Chung-Che
%S Proceedings of the 37th Conference on Computational Linguistics and Speech Processing (ROCLING 2025)
%D 2025
%8 November
%I Association for Computational Linguistics
%C National Taiwan University, Taipei City, Taiwan
%@ 979-8-89176-379-1
%F lee-etal-2025-hakka
%X 本研究為參加 FSR-2025 客語語音辨識挑戰賽(Hakka ASR II)的技術報告,旨在推進客語自動語音辨識技術的發展。由於客語屬於低資源語言,且存在多種腔調,語音辨識面臨高度挑戰。我們以 Whisperlarge-v2 為骨幹模型,設計兩階段訓練流程:首先利用「Hakka Across Taiwan(HAT)」語料庫進行模型調適,以捕捉客語的一般聲學特徵;其次在賽事方提供的60 小時腔調語料上進行微調,以增強對目標資料的適應性。實驗發現,直接輸出客語漢字可達到良好的字錯率(CER),但由 於腔調差異與拼音規則變化多,拼音任務表現顯著下降。為解決此問題,我們以漢字模型的編碼器初始化拼音模型,並提出結合 RoBERTa 漢字轉拼音、腔調判斷與字典修正的後處理模組,期望可以在比賽中提升辨識的成效。
%U https://aclanthology.org/2025.rocling-main.53/
%P 454-458
Markdown (Informal)
[Hakka Speech Recognition with Whisper and Pinyin Post-processing for FSR-2025](https://aclanthology.org/2025.rocling-main.53/) (Lee et al., ROCLING 2025)
ACL