arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.14583cs.CLcs.SDeess.AS
收藏

利用预先训练的表示来改善对濒危语言未转录语音的访问

Leveraging pre-trained representations to improve access to untranscribed speech from endangered languages

Stanford University(斯坦福大学);ARC Centre of Excellence for the Dynamics of Language(ARC语言动态卓越中心);University of Groningen(格罗宁根大学);University of Queensland(昆士兰大学);University of Melbourne(墨尔本大学);Australian National University(澳大利亚国立大学);University of Sydney(悉尼大学);University of Freiburg(弗莱堡大学)

Nay San, Martijn Bartelds, Mitchell Browne, Lily Clifford, Fiona Gibson, John Mansfield, David Nash, Jane Simpson, Myfany Turpin, Maria Vollmer, Sasha Wilmoth, Dan Jurafsky

摘要:

像wav 2 vec 2.0这样的预训练语音表示是自动语音识别(ASR)的强大工具。然而,许多濒危语言缺乏足够的数据来预先训练这些模型,或者主要是口头方言,没有标准化的书写系统,无法进行微调。举例查询口语词检测(QbE-STD)通过定位口语查询词为迭代地索引未转录语音语料库提供了一种替代方案。使用数据从7澳大利亚土著语言和各种区域的荷兰语,所有这些都是濒危或脆弱的,我们表明,QbE-STD可以通过利用表示开发ASR(wav 2 vec 2.0:英语单语模型和XLSR 53多语言模型)得到改善。令人惊讶的是,英语模型在4个澳大利亚语言数据集上的表现优于多语言模型,这就提出了如何最佳地利用QbE-STD的自监督语音表示的问题。尽管如此,我们发现wav 2 vec 2.0表示(英语或XLSR 53)在我们濒危语言数据集上的最先进方法上提供了很大的改进(相对56-86%)。

英文摘要:

Pre-trained speech representations like wav2vec 2.0 are a powerful tool for automatic speech recognition (ASR). Yet many endangered languages lack sufficient data for pre-training such models, or are predominantly oral vernaculars without a standardised writing system, precluding fine-tuning. Query-by-example spoken term detection (QbE-STD) offers an alternative for iteratively indexing untranscribed speech corpora by locating spoken query terms. Using data from 7 Australian Aboriginal languages and a regional variety of Dutch, all of which are endangered or vulnerable, we show that QbE-STD can be improved by leveraging representations developed for ASR (wav2vec 2.0: the English monolingual model and XLSR53 multilingual model). Surprisingly, the English model outperformed the multilingual model on 4 Australian language datasets, raising questions around how to optimally leverage self-supervised speech representations for QbE-STD. Nevertheless, we find that wav2vec 2.0 representations (either English or XLSR53) offer large improvements (56-86% relative) over state-of-the-art approaches on our endangered language datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑