Investigating gated recurrent networks for speech synthesis
Authors:
- 1 March 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5140-5144
Abstract
Recently, recurrent neural networks (RNNs) as powerful sequence models have re-emerged as a potential acoustic model for statistical parametric speech synthesis (SPSS). The long short-term memory (LSTM) architecture is particularly attractive because it addresses the vanishing gradient problem in standard RNNs, making them easier to train. Although recent studies have demonstrated that LSTMs can achieve significantly better performance on SPSS than deep feedforward neural networks, little is known about why. Here we attempt to answer two questions: a) why do LSTMs work well as a sequence model for SPSS; b) which component (e.g., input gate, output gate, forget gate) is most important. We present a visual analysis alongside a series of experiments, resulting in a proposal for a simplified architecture. The simplified architecture has significantly fewer parameters than an LSTM, thus reducing generation complexity considerably without degrading quality.Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- Modelling acoustic feature dependencies with artificial neural networks: Trajectory-RNADE2015
- Unidirectional long short-term memory recurrent neural network with recurrent output layer for low-latency speech synthesis2015
- Deep neural networks employing Multi-Task Learning and stacked bottleneck features for speech synthesis2015
- Deep mixture density networks for acoustic modeling in statistical parametric speech synthesis2014
- On the training aspects of Deep Neural Network (DNN) for parametric TTS synthesis2014
- Statistical parametric speech synthesisSpeech Communication, 2009
- Speech parameter generation algorithms for HMM-based speech synthesis2002
- Restructuring speech representations using a pitch-adaptive time–frequency smoothing and an instantaneous-frequency-based F0 extraction: Possible role of a repetitive structure in soundsSpeech Communication, 1999
- An RNN-based prosodic information synthesizer for Mandarin text-to-speechIEEE Transactions on Speech and Audio Processing, 1998
- Unit selection in a concatenative speech synthesis system using a large speech database1996