lmnt-com/diffwave

DiffWave is a fast, high-quality neural vocoder and waveform synthesizer.

48
/ 100
Emerging

DiffWave helps you convert complex sound representations, like Mel spectrograms, into realistic human speech or other audio waveforms. You provide either Gaussian noise or a conditioned signal, and it generates high-quality audio. This tool is ideal for audio engineers, researchers, or developers creating applications that need to generate natural-sounding voice from abstract audio features.

885 stars. No commits in the last 6 months.

Use this if you need to synthesize high-quality speech or other audio waveforms from structured inputs or even just from noise, and require fast, stable performance.

Not ideal if you're looking for a simple text-to-speech solution without needing to engage with spectrograms or audio feature engineering.

audio-synthesis speech-generation voice-AI audio-engineering digital-signal-processing
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

885

Forks

131

Language

Python

License

Apache-2.0

Last pushed

Mar 26, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/lmnt-com/diffwave"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.