no code implementations • 20 Jun 2023 • Jakub Swiatkowski, Duo Wang, Mikolaj Babianski, Giuseppe Coccia, Patrick Lumban Tobing, Ravichander Vipperla, Viacheslav Klimkov, Vincent Pollet
Speech generation for machine dubbing adds complexity to conventional Text-To-Speech solutions as the generated output is required to match the expressiveness, emotion and speaking rate of the source content.
no code implementations • 20 Jun 2023 • Jakub Swiatkowski, Duo Wang, Mikolaj Babianski, Patrick Lumban Tobing, Ravichander Vipperla, Vincent Pollet
Prosody transfer is well-studied in the context of expressive speech synthesis.
no code implementations • 13 Jul 2022 • Yi-Chiao Wu, Patrick Lumban Tobing, Kazuki Yasuhara, Noriyuki Matsunaga, Yamato Ohtani, Tomoki Toda
Neural-based text-to-speech (TTS) systems achieve very high-fidelity speech generation because of the rapid neural network developments.
2 code implementations • 20 May 2021 • Patrick Lumban Tobing, Tomoki Toda
To accommodate LLRT constraint with CPU, we propose a novel CycleVAE framework that utilizes mel-spectrogram as spectral features and is built with a sparse network architecture.
1 code implementation • 20 May 2021 • Patrick Lumban Tobing, Tomoki Toda
This paper presents a novel high-fidelity and low-latency universal neural vocoder framework based on multiband WaveRNN with data-driven linear prediction for discrete waveform modeling (MWDLP).
1 code implementation • 4 Mar 2021 • Kazuhiro Kobayashi, Wen-Chin Huang, Yi-Chiao Wu, Patrick Lumban Tobing, Tomoki Hayashi, Tomoki Toda
In this paper, we present an open-source software for developing a nonparallel voice conversion (VC) system named crank.
no code implementations • 9 Oct 2020 • Wen-Chin Huang, Patrick Lumban Tobing, Yi-Chiao Wu, Kazuhiro Kobayashi, Tomoki Toda
In this paper, we present the voice conversion (VC) systems developed at Nagoya University (NU) for the Voice Conversion Challenge 2020 (VCC2020).
1 code implementation • 9 Oct 2020 • Patrick Lumban Tobing, Yi-Chiao Wu, Tomoki Toda
In this paper, we present a description of the baseline system of Voice Conversion Challenge (VCC) 2020 with a cyclic variational autoencoder (CycleVAE) and Parallel WaveGAN (PWG), i. e., CycleVAEPWG.
1 code implementation • 11 Jul 2020 • Yi-Chiao Wu, Tomoki Hayashi, Patrick Lumban Tobing, Kazuhiro Kobayashi, Tomoki Toda
In this paper, a pitch-adaptive waveform generative model named Quasi-Periodic WaveNet (QPNet) is proposed to improve the limited pitch controllability of vanilla WaveNet (WN) using pitch-dependent dilated convolution neural networks (PDCNNs).
2 code implementations • 24 Jul 2019 • Patrick Lumban Tobing, Yi-Chiao Wu, Tomoki Hayashi, Kazuhiro Kobayashi, Tomoki Toda
In this work, to overcome this problem, we propose to use CycleVAE-based spectral model that indirectly optimizes the conversion flow by recycling the converted features back into the system to obtain corresponding cyclic reconstructed spectra that can be directly optimized.
1 code implementation • 21 Jul 2019 • Yi-Chiao Wu, Patrick Lumban Tobing, Tomoki Hayashi, Kazuhiro Kobayashi, Tomoki Toda
However, because of the fixed dilated convolution and generic network architecture, the WN vocoder lacks robustness against unseen input features and often requires a huge network size to achieve acceptable speech quality.
Audio and Speech Processing Sound
1 code implementation • 1 Jul 2019 • Yi-Chiao Wu, Tomoki Hayashi, Patrick Lumban Tobing, Kazuhiro Kobayashi, Tomoki Toda
In this paper, we propose a quasi-periodic neural network (QPNet) vocoder with a novel network architecture named pitch-dependent dilated convolution (PDCNN) to improve the pitch controllability of WaveNet (WN) vocoder.
1 code implementation • 2 May 2019 • Wen-Chin Huang, Yi-Chiao Wu, Chen-Chou Lo, Patrick Lumban Tobing, Tomoki Hayashi, Kazuhiro Kobayashi, Tomoki Toda, Yu Tsao, Hsin-Min Wang
Such hypothesis implies that during the conversion phase, the latent codes and the converted features in VAE based VC are in fact source F0 dependent.
no code implementations • 27 Nov 2018 • Wen-Chin Huang, Yi-Chiao Wu, Hsin-Te Hwang, Patrick Lumban Tobing, Tomoki Hayashi, Kazuhiro Kobayashi, Tomoki Toda, Yu Tsao, Hsin-Min Wang
Conventional WaveNet vocoders are trained with natural acoustic features but conditioned on the converted features in the conversion stage for VC, and such a mismatch often causes significant quality and similarity degradation.