法语神经 TTS:使用 SynPaFlex-Corpus 和 Tacotron2 比较字素输入与音素输入
Neural TTS in French: Comparing Graphemic and Phonetic Inputs Using the SynPaFlex-Corpus and Tacotron2
AI总结:
本文利用 SynPaFlex-Corpus 比较了字素与音素输入训练的法语单说话人 Tacotron2 模型,发现音素输入在发音准确度和韵律自然度上显著更优,且 6.1% 的音素错误率足以改善发音控制。
摘要:
SynPaFlex-Corpus 是一个公开可用的面向 TTS 的数据集,提供由 JTrans 转写器自动生成的音素转写,其音素错误率(PER)为 6.1%。在本文中,我们分析了两个基于 SynPaFlex-Corpus 提供的字素输入和音素输入训练的单说话人 Tacotron2 模型。通过三项主观听音测试,我们比较了它们的发音准确度、音质和自然度。我们的结果表明,基于音素的模型在发音准确度和韵律自然度上显著更好,但在感知音质方面没有显著差异。这些结果表明,在使用 83 小时录制的法语朗读语音时,6.1% 的 PER 足以通过使用音素转写而非字素来增强发音控制。它们还表明,SynPaFlex-Corpus 适合在单说话人微调方法中用于模型预训练。
英文摘要:
The SynPaFlex-Corpus is a publicly available TTS-oriented dataset, which provides phonetic transcriptions automatically produced by the JTrans transcriber, with a Phoneme Error Rate (PER) of 6.1%. In this paper, we analyze two mono-speaker Tacotron2 models trained on graphemic and phonetic inputs, provided by the SynPaFlex-Corpus. Through three subjective listening tests, we compare their pronunciation accuracy, sound quality and naturalness. Our results show significantly better pronunciation accuracy and prosody naturalness for the phoneme-based model, but no significant difference in terms of perceived sound quality. They demonstrate that a PER of 6.1% is sufficient to enhance pronunciation control by using phonetic transcripts instead of graphemes with 83 hours of recorded French read speech. They suggest that the SynPaFlex-Corpus is suitable for pre-training a model in mono-speaker fine-tuning approaches.
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。