The NGC catalog offers 100s of pre-trained models for computer vision, speech, recommendation, and more. Bring AI faster to market by using these models as-is or quickly build proprietary models with a fraction of your custom data.
This collection contains two models. 1) Multi-speaker 44100Hz FastPitch trained on approximately 20 hours of Latin American Spanish speech from 174 speakers. 2) HiFiGAN trained on mel spectrograms produced by the Multi-speaker FastPitch in (1).
For each word in the input text, the model: 1) predicts a punctuation mark that should follow the word (if any), the model supports commas, periods and question marks) and 2) predicts if the word should be capitalized or not.
HifiGAN is a neural vocoder model for text-to-speech applications. It is intended as the second part of a two-stage speech synthesis pipeline, with a mel-spectrogram generator such as FastPitch as the first stage.