Investigating Gated Recurrent Neural Networks For Speech Synthesis

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 feed-forward 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.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keywu2016investigating

Related papers

• Fast, Compact, And High Quality LSTM-RNN Based Statistical Parametric Speech Synthesizers For Mobile Devices (2016)13.05
• Memory Visualization For Gated Recurrent Neural Networks In Speech Recognition (2016)11.76
• Improving Speech Recognition By Revising Gated Recurrent Units (2017)11.19
• Light Gated Recurrent Units For Speech Recognition (2018)18.90
• Rnn-based Speech Synthesis Using A Continuous Sinusoidal Model (2019)3.58
• Generative Adversarial Network-based Glottal Waveform Model For Statistical Parametric Speech Synthesis (2019)10.35
• Utterance-level Sequential Modeling For Deep Gaussian Process Based Speech Synthesis Using Simple Recurrent Unit (2020)4.52
• Deep Feed-forward Sequential Memory Networks For Speech Synthesis (2018)5.84