MSc thesis on applying Deep Learning to speech synthesis, presented in Telecom Barcelona on July 2016.
- State of the art in Statistical Parametric Speech Synthesis
- Intro to Deep Learning
- Two stage TTS with RNN-LSTM and post-filtering
- Multiple Output Acoustic Mapping
- Multiple speakers acoustic shared representation
- Speaker adaptation
- Speaker interpolation
Please cite this work if it is useful for your research:
@mastersthesis{pascual2016deep,
title={Deep learning applied to speech synthesis},
author={Pascual de la Puente, Santiago},
year={2016},
school={Universitat Polit{\`e}cnica de Catalunya}
}
Santiago Pascual (@santty128)