gans   346

« earlier    

[1810.10989] Reducing over-smoothness in speech synthesis using Generative Adversarial Networks
Speech synthesis is widely used in many practical applications. In recent years, speech synthesis technology has developed rapidly. However, one of the reasons why synthetic speech is unnatural is that it often has over-smoothness. In order to improve the naturalness of synthetic speech, we first extract the mel-spectrogram of speech and convert it into a real image, then take the over-smooth mel-spectrogram image as input, and use image-to-image translation Generative Adversarial Networks(GANs) framework to generate a more realistic mel-spectrogram. Finally, the results show that this method greatly reduces the over-smoothness of synthesized speech and is more close to the mel-spectrogram of real speech.
sound  speech-synthesis  GANs  generative-models  texture  timbre  rather-interesting  uncanny-valley  feature-extraction  performance-measure  analytic-models-made-rough-again 
5 weeks ago by Vaguery
Twitter
We are open-sourcing VeGANs, a small library to easily train various existing using .

You provide a…
GANs  from twitter_favs
7 weeks ago by hustwj

« earlier    

related tags

2018  3d  ai  aicreated  analytic-models-made-rough-again  animation  anime  art  biggan  code  collaborative  comic  comics  computer-vision  conversion  creativity  crypto  culture  deep-learning  deep_learning  deepfakes  deeplearning  demos  densepose  design  dl  edwarddebelamy  example  face  faces  faking  feature-extraction  gallery  game-theory  gan  gen_art  gen_models  generation  generative-models  generative  generators  github  graphics  ifttt  images  inspiration  inverse  journalism  language  lol  machine-learning  machinelearning  maps  mario-klingemann  material  math  media  molecular  neural-networks  neuralnets  neuralnetworks  nlp  nvidia  obvious  online  opencv  performance-measure  pics2pics  pix2pix  rather-interesting  regression  scene  sothebys  sound  speech-synthesis  style-transfer  styletransfer  sweet  teaching  tech  tensorflow  texture  thenetwork  timbre  time-series  tutorial  tutorials  uncanny-valley  vae  video  visual  visualization 

Copy this bookmark:



description:


tags: