babysor/MockingBird

🚀Clone a voice in 5 seconds to generate arbitrary speech in real-time

58
/ 100
Established

Uses a modular three-stage architecture with pretrained speaker encoder and neural vocoder, training only a Mandarin-optimized synthesizer to reduce computational overhead. Operates as both a PyQt5 desktop toolbox and web server, supporting inference on GPU (CUDA) and CPU across Windows, Linux, and M1 Mac via Rosetta emulation. Extensively tested on Chinese speech datasets (aidatatang_200zh, aishell3, magicdata) with PyTorch 1.9.0+, allowing users to train custom synthesizers or leverage community pretrained models.

36,874 stars.

No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

36,874

Forks

5,236

Language

Python

License

Last pushed

Mar 03, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/babysor/MockingBird"

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