Tacotron 2

GPTKB entity

Statements (48)
Predicate Object
gptkbp:instanceOf speech synthesis system
gptkbp:application text-to-speech
gptkbp:arXivID 1712.05884
gptkbp:author gptkb:Navdeep_Jaitly
gptkb:Mike_Schuster
gptkb:RJ_Skerry-Ryan
gptkb:Rif_A._Saurous
gptkb:Ron_J._Weiss
gptkb:Yannis_Agiomyrgiannakis
gptkb:Yonghui_Wu
gptkb:Zhifeng_Chen
gptkb:Zongheng_Yang
Yuxuan Wang
Jonathan Shen
Ruoming Pang
Yu Zhang
gptkbp:category gptkb:artificial_intelligence
gptkb:public_speaker
deep learning
gptkbp:citation 2017
Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
gptkbp:component gptkb:convolutional_neural_network
recurrent neural network
attention mechanism
gptkbp:developedBy gptkb:Google
gptkbp:github https://github.com/NVIDIA/tacotron2
https://www.w3.org/2000/01/rdf-schema#label Tacotron 2
gptkbp:influenced gptkb:Glow-TTS
FastSpeech
Parallel Tacotron
gptkbp:input gptkb:text
character sequence
gptkbp:language English
gptkbp:notableFor end-to-end text-to-speech synthesis
high-quality natural speech
gptkbp:openSource yes
gptkbp:output speech waveform
gptkbp:outputRepresentation mel spectrogram
gptkbp:predecessor gptkb:Tacotron
gptkbp:publishedIn gptkb:arXiv
gptkbp:releaseYear 2017
gptkbp:trainer LJSpeech
VCTK
paired text and speech
gptkbp:uses sequence-to-sequence model
WaveNet vocoder
gptkbp:bfsParent gptkb:Tacotron
gptkbp:bfsLayer 7