WaveGlow: flow-based parallel mel-spectrogram vocoding

What changed

Combined Glow-style invertible transformations and WaveNet-derived components in a likelihood-trained parallel waveform synthesizer conditioned on mel spectrograms.

Scientific scope

Combined Glow-style invertible transformations and WaveNet-derived components in a likelihood-trained parallel waveform synthesizer conditioned on mel spectrograms. Inherits normalizing flows and WaveNet. Parallel synthesis is not first non-autoregressive vocoding; the model is a waveform stage rather than complete raw-text speech system. Source throughput/MOS not independently replayed.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Inherits normalizing flows and WaveNet. Parallel synthesis is not first non-autoregressive vocoding; the model is a waveform stage rather than complete raw-text speech system. Source throughput/MOS not independently replayed.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• WaveGlow: A Flow-based Generative Network for Speech Synthesis

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction