Area of research
Artificial Intelligence · Signal Processing
Research interest
Research focused on Speech recognition and End-to-end principle, with related work in Beamforming, Spectrogram, Transformer. Notable publications include 'Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesis', 'Deep beamforming networks for multi-channel speech recognition', and 'Espnet-TTS: Unified, Reproducible, and Integratable Open Source End-to-End Text-to-Speech Toolkit'.
Espnet-TTS: Unified, Reproducible, and Integratable Open Source End-to-End Text-to-Speech Toolkit
Disentangling Correlated Speaker and Noise for Speech Synthesis via Data Augmentation and Adversarial Factorization
Semi-supervised Training for Improving Data Efficiency in End-to-end Speech Synthesis
Transfer Learning from Speaker Verification to Multispeaker Text-To-Speech Synthesis
Neural Information Processing Systems 2018cited by 230position: middle
Deep beamforming networks for multi-channel speech recognition