2026-04-20T12:01:30Z https://ipsj.ixsq.nii.ac.jp/oai

oai:ipsj.ixsq.nii.ac.jp:00185799 2025-01-20T02:49:42Z 1164:5159:9402:9403

Investigation of WaveNet for Text-to-Speech Synthesis Investigation of WaveNet for Text-to-Speech Synthesis Xin, Wang Shinji, Takaki Junichi, Yamagishi Xin, Wang Shinji, Takaki Junichi, Yamagishi WaveNet is a type of neural network that can be used to model speech waveforms. It has been used in text-to-speech synthesis systems to convert acoustic or linguistic features into waveforms. Despite the description in recent literatures and open-source implementation, the mechanism of WaveNet is still somewhat obscure. This work explains the authors' WaveNet implementation. It also introduces a one-best generation method that could be an alternative to the random-sampling-based generation method. Based on the implementation, this work shows observations inside the network. Interesting findings include the manifold of quantized waveforms learned by WaveNet and the gradually decreased data variance in WaveNet blocks. These results may be helpful for further investigation on WaveNet. WaveNet is a type of neural network that can be used to model speech waveforms. It has been used in text-to-speech synthesis systems to convert acoustic or linguistic features into waveforms. Despite the description in recent literatures and open-source implementation, the mechanism of WaveNet is still somewhat obscure. This work explains the authors' WaveNet implementation. It also introduces a one-best generation method that could be an alternative to the random-sampling-based generation method. Based on the implementation, this work shows observations inside the network. Interesting findings include the manifold of quantized waveforms learned by WaveNet and the gradually decreased data variance in WaveNet blocks. These results may be helpful for further investigation on WaveNet. technical report 情報処理学会 2018-02-13 application/pdf 研究報告音声言語情報処理（SLP） 6 2018-SLP-120 1 6 2188-8663 AN10442647 https://ipsj.ixsq.nii.ac.jp/record/185799/files/IPSJ-SLP18120006.pdf eng