Non-Attentive Tacotron: Robust and Controllable Neural TTS Synthesis Including Unsupervised Duration Modeling

This paper presents Non-Attentive Tacotron based on the Tacotron 2 text-to-speech model, replacing the attention mechanism with an explicit duration predictor. This improves robustness significantly as measured by unaligned duration ratio and word deletion rate, two metrics introduced in this paper for large-scale robustness evaluation using a pre-trained speech recognition model... (read more)

PDF Abstract
No code implementations yet. Submit your code now

Datasets


Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods used in the Paper


METHOD TYPE
Highway Layer
Miscellaneous Components
Max Pooling
Pooling Operations
Dilated Causal Convolution
Temporal Convolutions
Tanh Activation
Activation Functions
Highway Network
Feedforward Networks
Sigmoid Activation
Activation Functions
Residual GRU
Recurrent Neural Networks
Residual Connection
Skip Connections
Griffin-Lim Algorithm
Phase Reconstruction
BiGRU
Bidirectional Recurrent Neural Networks
CBHG
Speech Synthesis Blocks
LSTM
Recurrent Neural Networks
Convolution
Convolutions
Additive Attention
Attention Mechanisms
GRU
Recurrent Neural Networks
Dense Connections
Feedforward Networks
Tacotron
Text-to-Speech Models
BiLSTM
Bidirectional Recurrent Neural Networks
WaveNet
Generative Audio Models
Linear Layer
Feedforward Networks
Zoneout
Regularization
Batch Normalization
Normalization
Dropout
Regularization
Mixture of Logistic Distributions
Output Functions
ReLU
Activation Functions
Location Sensitive Attention
Attention Mechanisms
Tacotron 2
Text-to-Speech Models