Qwen3 TTS reference voices
Twenty synthetic reference voices (one female and one male in ten languages) generated with Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign, dedicated to the public domain under CC0 1.0, for voice cloning with Qwen3 TTS, VoxCPM2 and Chatterbox. See NOTICE.md and voices.json.
Also available at https://github.com/exnihilus/subvoxpro-models/tree/main/qwen3-voices.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for ExNihilus/qwen3-tts-reference-voices
Base model
Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign