FastSpeech

E1866721 UNEXPLORED

FastSpeech is a non-autoregressive text-to-speech model that generates speech quickly and robustly by predicting mel-spectrograms in parallel rather than step-by-step.

Try in SPARQL Jump to: Surface forms Referenced by

All labels observed (1)

Label Occurrences
FastSpeech canonical 1

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

Tacotron influenced FastSpeech