BREAKING
LuxTTS: Voice Cloning in 1GB VRAM
0x
realtime on GPU
0GB
VRAM footprint
0kHz
audio output
How LuxTTS Was Built
1Based on ZipVoice
2Distilled to 4 steps
348kHz vocoder
4Zero-shot cloning
0s
reference audio
0MB
min model size
0MB
max model size
Local Alternative to Cloud TTS
LuxTTSApache-2.0
Runs in 1GB VRAM
CUDA, CPU and MPS
Free and open source
Commercial TTS
Cloud-dependent
Paid access
Less local control
Small Model, Big Reach
AI NEWS BLITZ
A new open-source TTS model clones voices locally in just one gigabyte of VRAM.