Open-source model with complete transparency and flexibility
Trained on 10M hours of audio data for superior voice quality
Real-time low latency inference
Supports WAV, MP3, and FLAC formats
14-day free trial for all paid plans
FAQ
How does Higgs Audio v2 work?: Uses advanced neural networks trained on 10M hours of audio data, simply provide text and optional voice reference for cloning
Can I upgrade my plan anytime?: Yes, upgrade or downgrade anytime, changes take effect immediately with prorated billing
What audio formats are supported?: WAV, MP3, and FLAC formats in 24kHz high-fidelity quality
Is there a free trial?: All paid plans come with a 14-day free trial