vocence-tts-v8

Prompt-driven English TTS based on Qwen3-TTS-12Hz-1.7B-VoiceDesign with self-imitation SFT.

from qwen_tts import Qwen3TTSModel
m = Qwen3TTSModel.from_pretrained("artur7236/vocence-tts-v8")
wavs, sr = m.generate_voice_design(text="Hi.", instruct="Friendly voice.", language="english")

CC BY-NC-SA 4.0.

Downloads last month: 4

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for artur7236/vocence-tts-v18

Base model

Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign

Finetuned

(40)

this model