paper-with-me

Papers

Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions

2017-12-16 · Jonathan Shen, Ruoming Pang, Ron J. Weiss, Mike Schuster, Navdeep Jaitly, Zongheng Yang, Zhifeng Chen, Yu Zhang, Yuxuan Wang, RJ Skerry-Ryan, Rif A. Saurous, Yannis Agiomyrgiannakis, Yonghui Wu

This paper describes Tacotron 2, a neural network architecture for speech synthesis directly from text. The system is composed of a recurrent sequence-to-sequence feature prediction network that maps character embeddings to mel-scale spectrograms, followed by a modified WaveNet model acting as a vocoder to synthesize timedomain waveforms from those spectrograms. Our model achieves a mean opinion score (MOS) of $4.53$ comparable to a MOS of $4.58$ for professionally recorded speech. To validate our design choices, we present ablation studies of key components of our system and evaluate the impact of using mel spectrograms as the input to WaveNet instead of linguistic, duration, and $F_0$ features. We further demonstrate that using a compact acoustic intermediate representation enables significant simplification of the WaveNet architecture.

📄 PDF Abstract BibTeX arXiv:1712.05884

Code (33)

BogiHsu/Tacotron2-PyTorch pytorch
Jeevesh8/Cross-Lingual-Voice-Cloning pytorch
MindSpore-paper-code-2/code400/tree/main/Tacotron2 mindspore
NVIDIA/tacotron2 pytorch
OlaWod/my-tacotron2 pytorch
PaddlePaddle/PaddleSpeech paddle
Rayhane-mamah/Tacotron-2 tf
TensorSpeech/TensorflowTTS tf
alpharol/Taco_Collection tf
anandaswarup/TTS pytorch
anandaswarup/rnn-tts pytorch
bfs18/tacotron2 pytorch
choiHkk/Transformer-TTS pytorch
codetendolkar/tacotron-2-explained tf
coqui-ai/TTS pytorch
creotiv/RussianTTS-Tacotron2 pytorch
dathudeptrai/TensorflowTTS tf
dipjyoti92/SC-WaveRNN pytorch
dipjyoti92/TTS-Style-Transfer pytorch
izzajalandoni/tts_models pytorch
jiean001/models_m/tree/main/tacotron2 mindspore
kaiidams/voice100 pytorch
kaiidams/voice100-tts pytorch
keonlee9420/Comprehensive-Tacotron2 pytorch
martinlenglet/avtacotron2 pytorch
mindspore-ai/models/tree/master/research/audio/tacotron2 mindspore
rosinality/melgan-pytorch pytorch
s3nh/pytorch-tacotron2 pytorch
thepowerfuldeez/tacotron2 pytorch
thuhcsi/tacotron pytorch
vincenzo-scotti/tacotron2 pytorch
xcmyz/FastSpeech pytorch
xinshengwang/Tacotron-pytorch pytorch

Tasks

Speech Synthesis

Methods 이 논문이 사용한 방법론

Zoneout Zoneout is a method for regularizing RNNs. At each timestep, zoneout stochastically forces some…
LSTM An LSTM is a type of recurrent neural network that addresses the vanishing gradient problem in vanilla…
Mixture of Logistic Distributions 설명 없음
Location Sensitive Attention Location Sensitive Attention is an attention mechanism that extends the additive attention mechanism to use cumulative…
BiLSTM A Bidirectional LSTM, or biLSTM, is a sequence processing model that consists of two LSTMs: one taking the input in a forward direction, and the other in a backwards…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Exponential Decay Exponential Decay is a learning rate schedule where we decay the learning rate with more iterations using an exponential function: $$ \text{lr} =…
Weight Decay 설명 없음

Similar Papers 제목 키워드 기반

Parallel WaveNet conditioned on VAE latent vectors

2020-12-17 · Jonas Rohnke, Tom Merritt, Jaime Lorenzo-Trueba, Adam Gabrys 외

Recently the state-of-the-art text-to-speech synthesis systems have shifted to a two-model approach: a sequence-to-sequence model to predict a representation of speech (typically mel-spectrograms), followed by a 'neural …

SentenceSpeech Synthesistext-to-speechText to Speech+1

GELP: GAN-Excited Linear Prediction for Speech Synthesis from Mel-spectrogram

2019-04-08 · Lauri Juvela, Bajibabu Bollepalli, Junichi Yamagishi, Paavo Alku

Recent advances in neural network -based text-to-speech have reached human level naturalness in synthetic speech. The present sequence-to-sequence models can directly map text to mel-spectrogram acoustic features, which …

Speech Synthesistext-to-speechText to Speech

Parametric Resynthesis with neural vocoders

2019-06-16 · Soumi Maiti, Michael I Mandel

Noise suppression systems generally produce output speech with compromised quality. We propose to utilize the high quality speech generation capability of neural vocoders for noise suppression. We use a neural network to…

Resynthesis

High-quality Speech Synthesis Using Super-resolution Mel-Spectrogram

2019-12-03 · Leyuan Sheng, Dong-Yan Huang, Evgeniy N. Pavlovskiy

In speech synthesis and speech enhancement systems, melspectrograms need to be precise in acoustic representations. However, the generated spectrograms are over-smooth, that could not produce high quality synthesized spe…

Image-to-Image TranslationSpeech EnhancementSpeech SynthesisSuper-Resolution+2

Towards Robust FastSpeech 2 by Modelling Residual Multimodality

2023-06-02 · Fabian Kögel, Bac Nguyen, Fabien Cardinaux

State-of-the-art non-autoregressive text-to-speech (TTS) models based on FastSpeech 2 can efficiently synthesise high-fidelity and natural speech. For expressive speech datasets however, we observe characteristic audio d…

Decodertext-to-speechText to Speech