Modeling spectral envelopes using restricted Boltzmann machines for statistical parametric speech synthesis

By Zhenhua Ling, Li Deng, Dong Yu. 2013

52 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Zhenhua Ling
2. Frank K. Soong
3. Junichi Yamagishi
4. Li Deng
5. Dong Yu
6. Lirong Dai
7. Yao Qian
8. Yuchen Fan

In Signal Processing

1. Li Deng
2. Dong Yu
3. Frank K. Soong
4. Lirong Dai
5. Zhenhua Ling
6. Yuchen Fan
7. Yao Qian
8. Jinyu Li

Citations in the corpus

• Deep Learning: Methods and Applications (2014, background): “Similar trajectory HMMs also form the basis for parametric speech synthesis [228, 326, 439, 438].”
• Statistical parametric speech synthesis using generative adversarial networks under a multi-task learning framework (2017, unknown): “More recently, advanced estimation criteria and novel network architectures have been introduced to further improve the performance of SPSS [7, 8, 9, 10, 11].”
• Voice Conversion Using Deep Neural Networks With Layer-Wise Generative Training (2014, background): “The source features that belongs to thesamemixture thus tend to be converted to the same target, which is a weighted average of means of the target GMM model.”
• A Review of Deep Learning Based Speech Synthesis (2019, methodology): “It is adopted to better describe the distribution of high-dimensional spectral envelopes to alleviate the over-smooth problem in HMM-based speech synthesis [40].”
• Modeling Spectral Envelopes Using Restricted Boltzmann Machines and Deep Belief Networks for Statistical Parametric Speech Synthesis (2013, methodology): “metric speech synthesis have also been studied very recently [1],”
• TTS synthesis with bidirectional LSTM based recurrent neural networks (2014, methodology): “In addition, RBM is directly used to represent the distribution of the spectral envelops at each HMM state in [8], where the mode of RBM has been shown to perform better than the mean of GMM and results in better synthesized voice quality.”