AIConfPaper
← Search

Yoshinori Shiga

7 accepted papers

2021

High-Intelligibility Speech Synthesis for Dysarthric Speakers with LPCNet-Based TTS and CycleVAE-Based VC

Keisuke Matsubara, Takuma Okamoto, Ryoichi Takashima, Tetsuya Takiguchi, Tomoki Toda, Yoshinori Shiga +1
ICASSP 2021accepted

This paper presents a high-intelligibility speech synthesis method for persons with dysarthria caused by athetoid cerebral palsy. The muscular control of such speakers is unstable because of their athetoid symptoms, and their pronunciation is unclear, which makes it difficult for them to communicate…

Cited by 0Source ↗Scholar ↗
2021

Noise Level Limited Sub-Modeling for Diffusion Probabilistic Vocoders

Takuma Okamoto, Tomoki Toda, Yoshinori Shiga, Hisashi Kawai
ICASSP 2021accepted

Although diffusion probabilistic vocoders WaveGrad and DiffWave can realize real-time high-fidelity speech synthesis with a simple loss function in training, all noise components with over the full range of noise levels are predicted by one model in all iterations. This paper proposes a simple but e…

Cited by 0Source ↗Scholar ↗
2020

Transformer-Based Text-to-Speech with Weighted Forced Attention

Takuma Okamoto, Tomoki Toda, Yoshinori Shiga, Hisashi Kawai
ICASSP 2020accepted

This paper investigates state-of-the-art Transformer- and FastSpeech-based high-fidelity neural text-to-speech (TTS) with full-context label input for pitch accent languages. The aim is to realize faster training than conventional Tacotron-based models. Introducing phoneme durations into Tacotron-ba…

Cited by 0Source ↗Scholar ↗
2019

Investigations of Real-time Gaussian Fftnet and Parallel Wavenet Neural Vocoders with Simple Acoustic Features

Takuma Okamoto, Tomoki Toda, Yoshinori Shiga, Hisashi Kawai
ICASSP 2019accepted

This paper examines four approaches to improving real-time neural vocoders with simple acoustic features (SAF) constructed from fundamental frequency and mel-cepstra rather than mel-spectrograms. The investigations are as follows: 1) the effectiveness of single Gaussian (SG) autoregressive (AR) Wave…

Cited by 0Source ↗Scholar ↗
2018

An Investigation of Noise Shaping with Perceptual Weighting for Wavenet-Based Speech Generation

Kentaro Tachibana, Tomoki Toda, Yoshinori Shiga, Hisashi Kawai
ICASSP 2018accepted

We propose a noise shaping method to improve the sound quality of speech signals generated by WaveNet, which is a convolutional neural network (CNN) that predicts a waveform sample sequence as a discrete symbol sequence. Speech signals generated by WaveNet often suffer from noise signals caused by t…

Cited by 0Source ↗Scholar ↗
2018

An Investigation of Subband Wavenet Vocoder Covering Entire Audible Frequency Range with Limited Acoustic Features

Takuma Okamoto, Kentaro Tachibana, Tomoki Toda, Yoshinori Shiga, Hisashi Kawai
ICASSP 2018accepted

Although a WaveNet vocoder can synthesize more natural-sounding speech waveforms than conventional vocoders with sampling frequencies of 16 and 24 kHz, it is difficult to directly extend the sampling frequency to 48 kHz to cover the entire human audible frequency range for higher-quality synthesis b…

Cited by 0Source ↗Scholar ↗