rishikksh20/UnivNet-pytorch

UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generation

38
/ 100
Emerging

This project helps audio engineers and researchers generate highly realistic, natural-sounding speech from spectrograms. You input a spectrogram, which is a visual representation of the frequencies in an audio signal, and it outputs a high-fidelity audio waveform. It's designed for professionals working on voice synthesis, text-to-speech systems, or audio manipulation.

No commits in the last 6 months.

Use this if you need to transform spectrograms into high-quality, lifelike audio, especially for speech synthesis applications.

Not ideal if you are looking for a complete, end-to-end voice cloning or text-to-speech toolbox without needing to interact with spectrograms directly.

audio-synthesis voice-generation spectrogram-to-audio neural-vocoders speech-technology
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

76

Forks

9

Language

Python

License

MIT

Last pushed

Aug 30, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/rishikksh20/UnivNet-pytorch"

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