arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.06372cs.CLcs.SDeess.AS
收藏

SpeechTaxi:论多语言语义语音分类

SpeechTaxi: On Multilingual Semantic Speech Classification

Center For Artificial Intelligence and Data Science, University of Würzburg(维尔茨堡大学人工智能与数据科学中心)

Lennart Keller, Goran Glavaš

AI总结:

本文构建覆盖28种语言、时长80小时的SpeechTaxi数据集,系统比较端到端与级联语义语音分类,并提出罗马化文本级联方法以解决低资源语言ASR支持不足问题。

摘要:

多语言语音编码和转录的最新进展提出了语义语音分类最有效方法的问题。具体而言,(1)通过微调最先进的多语言语音编码器(MSE)获得的端到端(E2E)分类器,能否匹敌或超过(2)级联(CA)方法的性能?在CA中,语音首先被转录为文本,分类则交由基于文本的分类器完成。为回答这一问题,我们首先构建了SpeechTaxi,这是一个用于圣经经文语义语音分类的80小时多语言数据集,覆盖28种不同语言。随后,我们利用SpeechTaxi开展大量实验,在单语言语义语音分类以及跨语言迁移中比较E2E与CA。我们发现,在单语言设置下,即在语言内数据上训练时,基于MSE的E2E优于CA。然而,MSE的跨语言迁移能力似乎较差:无论是在(1)向训练中未见语言的零样本迁移,还是在(2)多语言训练(即在多种语言上联合训练)中,E2E都显著落后于CA。最后,我们设计了一种新颖的CA方法,该方法基于将语音转录为罗马化文本,并将其作为语言无关的中间表示;结果表明,对于缺乏原生ASR支持的语言,这是一种稳健的解决方案。我们的SpeechTaxi数据集公开发布于:https://huggingface.co/datasets/LennartKeller/SpeechTaxi/。

英文摘要:

Recent advancements in multilingual speech encoding as well as transcription raise the question of the most effective approach to semantic speech classification. Concretely, can (1) end-to-end (E2E) classifiers obtained by fine-tuning state-of-the-art multilingual speech encoders (MSEs) match or surpass the performance of (2) cascading (CA), where speech is first transcribed into text and classification is delegated to a text-based classifier. To answer this, we first construct SpeechTaxi, an 80-hour multilingual dataset for semantic speech classification of Bible verses, covering 28 diverse languages. We then leverage SpeechTaxi to conduct a wide range of experiments comparing E2E and CA in monolingual semantic speech classification as well as in cross-lingual transfer. We find that E2E based on MSEs outperforms CA in monolingual setups, i.e., when trained on in-language data. However, MSEs seem to have poor cross-lingual transfer abilities, with E2E substantially lagging CA both in (1) zero-shot transfer to languages unseen in training and (2) multilingual training, i.e., joint training on multiple languages. Finally, we devise a novel CA approach based on transcription to Romanized text as a language-agnostic intermediate representation and show that it represents a robust solution for languages without native ASR support. Our SpeechTaxi dataset is publicly available at: https://huggingface.co/ datasets/LennartKeller/SpeechTaxi/.

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。

↑