A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit <a rel="external noopener" href="https://arxiv.org/pdf/2005.10390v2.pdf">the original URL</a>. The file type is <code>application/pdf</code>.
<span class="release-stage" >pre-print</span>
Neural sequence-to-sequence text-to-speech synthesis (TTS) can produce high-quality speech directly from text or simple linguistic features such as phonemes. Unlike traditional pipeline TTS, the neural sequence-to-sequence TTS does not require manually annotated and complicated linguistic features such as part-of-speech tags and syntactic structures for system training. However, it must be carefully designed and well optimized so that it can implicitly extract useful linguistic features from<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.10390v2">arXiv:2005.10390v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/bnny4jsnqvf4nl4gpeias5lbjm">fatcat:bnny4jsnqvf4nl4gpeias5lbjm</a> </span>
more »... input features. In this paper we investigate under what conditions the neural sequence-to-sequence TTS can work well in Japanese and English along with comparisons with deep neural network (DNN) based pipeline TTS systems. Unlike past comparative studies, the pipeline systems also use autoregressive probabilistic modeling and a neural vocoder. We investigated systems from three aspects: a) model architecture, b) model parameter size, and c) language. For the model architecture aspect, we adopt modified Tacotron systems that we previously proposed and their variants using an encoder from Tacotron or Tacotron2. For the model parameter size aspect, we investigate two model parameter sizes. For the language aspect, we conduct listening tests in both Japanese and English to see if our findings can be generalized across languages. Our experiments suggest that a) a neural sequence-to-sequence TTS system should have a sufficient number of model parameters to produce high quality speech, b) it should also use a powerful encoder when it takes characters as inputs, and c) the encoder still has a room for improvement and needs to have an improved architecture to learn supra-segmental features more appropriately.
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201011082801/https://arxiv.org/pdf/2005.10390v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.10390v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>