← Search

Aaron Oord

3 accepted papers

2018

Efficient Neural Audio Synthesis

ICML 2018oral

Sequential models achieve state-of-the-art results in audio, visual and textual domains with respect to both estimating the data distribution and generating desired samples. Efficient sampling for this class of models at the cost of little to no loss in quality has however remained an elusive proble…

Cited by 1097SourcePDFScholar
2018

Parallel WaveNet: Fast High-Fidelity Speech Synthesis

ICML 2018oral

The recently-developed WaveNet architecture is the current state of the art in realistic speech synthesis, consistently rated as more natural sounding for many different languages than any previous system. However, because WaveNet relies on sequential generation of one audio sample at a time, it is…

Cited by 1053SourcePDFScholar