Can We Train a Language Model Inside an End-to-End ASR Model? - Investigating Effective Implicit Language Modeling

no code implementations • CAI (COLING) 2022 • Zhuo Gong, Daisuke Saito, Sheng Li, Hisashi Kawai, Nobuaki Minematsu

The experiments show that we can enhance an ASR E2E model based on encoder-decoder architecture by pre-training the decoder with text data.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +2

Hierarchical Softmax for End-to-End Low-resource Multilingual Speech Recognition

1 code implementation • 8 Apr 2022 • Qianying Liu, Zhuo Gong, Zhengdong Yang, Yuhang Yang, Sheng Li, Chenchen Ding, Nobuaki Minematsu, Hao Huang, Fei Cheng, Chenhui Chu, Sadao Kurohashi

Low-resource speech recognition has been long-suffering from insufficient training data.

speech-recognition Speech Recognition

Wasserstein GAN and Waveform Loss-based Acoustic Model Training for Multi-speaker Text-to-Speech Synthesis Systems Using a WaveNet Vocoder

no code implementations • 31 Jul 2018 • Yi Zhao, Shinji Takaki, Hieu-Thi Luong, Junichi Yamagishi, Daisuke Saito, Nobuaki Minematsu

In order to reduce the mismatched characteristics between natural and generated acoustic features, we propose frameworks that incorporate either a conditional generative adversarial network (GAN) or its variant, Wasserstein GAN with gradient penalty (WGAN-GP), into multi-speaker speech synthesis that uses the WaveNet vocoder.

Generative Adversarial Network Speech Synthesis +1

WFST-Based Grapheme-to-Phoneme Conversion: Open Source tools for Alignment, Model-Building and Decoding

no code implementations • WS 2012 • Josef R. Novak, Nobuaki Minematsu, Keikichi Hirose

Language Modelling Speech Recognition