Demo page of an efficiently trainable TTS

• This is a demo page of the paper “Efficiently Trainable Text-to-Speech System Based on Deep Convolutional Networks With Guided Attention”, ICASSP 2018.
• Paper available at
  • IEEE Xplore
  • arXiv

@inproceedings{tachibana2018efficiently,
   author={Tachibana, H. and Uenoyama, K. and Aihara, S.},
   title={Efficiently Trainable Text-to-Speech System Based on Deep Convolutional Networks with Guided Attention},
   booktitle={2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
   publisher={IEEE},
   month={Apr},
   year={2018},
   DOI={10.1109/icassp.2018.8461829}
}

Samples

Long Sentences

icassp stands for the international conference on acoustics, speech and signal processing.　

Synthesized audio (15 hours training).

Note: the words “ICASSP” and “acoustics” are not included in our training data.

generative adversarial network or variational autoencoder.

a matrix is positive definite, if all eigenvalues are positive.

a matrix is positive semi-definite, if all eigenvalues are non-negative.

if the hessian matrix of a multivariate function is positive semi-definite, the function is convex.

there always exists at least one mixed strategy nash equilibrium.

a spectrogram is obtained by applying es-tee-ef-tee to a signal.

the two player zero-sum game of wasserstein gan is derived by considering kantorovich-rubinstein duality.

keras is a famous deep learning framework for python, which uses tensorflow as a backend.

solidity is a turing complete programming language for ethereum, which is the second largest cryptocurrency based on blockchain.

brahms’s symphony number one is sometimes called beethoven’s tenth symphony.

a british author, kazuo ishiguro, famous for his works such as the remains of the day, was awarded the nobel prize in literature in two thousand seventeen.

tokyo is the most populous megacity in the world, with nearly thirty-eight million population.

in the chinese language, the word, crisis, is composed of two characters, one representing danger and the other, opportunity. (John F. Kennedy)

Note: This famous quote is not included in the training data.

to be, or not to be, that is the question. (Shakespear)

Note: This famous quote is not included in the training data.