通过跨语言伪标签实现无监督ASR
Unsupervised ASR via Cross-Lingual Pseudo-Labeling
AI总结：
提出跨语言伪标签方法，利用其他语言的字符级声学模型和目标语言模型约束，在无目标语言标注音频下实现高效无监督ASR。
摘要：
近期工作表明，仅使用无配对的音频和文本即可训练无监督自动语音识别（ASR）系统。现有无监督ASR方法假设无标签数据可用于训练。我们论证，即使没有某语言的标注音频，其他语言始终存在可用的标注数据。我们证明，可利用其他语言的字符级声学模型（AM）来引导新语言的无监督AM。此处“无监督”指目标语言无可用标注音频。我们的方法基于两个关键要素：（i）使用其他语言AM生成目标语言的伪标签（PLs），以及（ii）使用目标语言模型约束这些PLs。该方法在Common Voice上有效：例如，将英语AM迁移至斯瓦希里语实现了18%的WER。在LJSpeech上，使用800小时标注德语数据替代60k小时无标注英语数据，它也以15%的绝对WER优势超越了基于字符的wav2vec-U 2.0。
英文摘要：
Recent work has shown that it is possible to train an $\textit{unsupervised}$ automatic speech recognition (ASR) system using only unpaired audio and text. Existing unsupervised ASR methods assume that no labeled data can be used for training. We argue that even if one does not have any labeled audio for a given language, there is $\textit{always}$ labeled data available for other languages. We show that it is possible to use character-level acoustic models (AMs) from other languages to bootstrap an $\textit{unsupervised}$ AM in a new language. Here, "unsupervised" means no labeled audio is available for the $\textit{target}$ language. Our approach is based on two key ingredients: (i) generating pseudo-labels (PLs) of the $\textit{target}$ language using some $\textit{other}$ language AM and (ii) constraining these PLs with a $\textit{target language model}$. Our approach is effective on Common Voice: e.g. transfer of English AM to Swahili achieves 18% WER. It also outperforms character-based wav2vec-U 2.0 by 15% absolute WER on LJSpeech with 800h of labeled German data instead of 60k hours of unlabeled English data.
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。
