Downloads · 30 days
232
25% of all-time downloads
Qwen3-TTS-12Hz-0.6B-CustomVoice-MLX-bf16
aufklarer/Qwen3-TTS-12Hz-0.6B-CustomVoice-MLX-bf16
Qwen3-TTS-12Hz-0.6B-CustomVoice-MLX-bf16 is a machine learning model from aufklarer. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as apache-2.0.
MLX bf16 (fp16) conversion of Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice for Apple Silicon inference. Weights are unquantized — the highest-fidelity CustomVoice variant.
All-time downloads
940
Public
Parameters
906M
1.8 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors1.8 GB · 100%
At a glance
- Library
- mlx
- License
- apache-2.0
- Model type
- qwen3_tts
- Access
- Public
- Created
- Jun 21, 2026
- Updated
- Jun 21, 2026
- SHA
- 3affbf65
Base models
Model card
From the Hugging Face model README
Qwen3 TTS 12Hz 0.6B CustomVoice — MLX bf16
MLX bf16 (fp16) conversion of Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice for Apple Silicon inference. Weights are unquantized — the highest-fidelity CustomVoice variant.
Usage
Used by speech-swift Qwen3TTS module:
let model = try await Qwen3TTSModel.fromPretrained(
modelId: TTSModelVariant.customVoice.rawValue
)
let audio = try model.synthesize("Hello!", speaker: "ryan")
audio speak "Hello!" --model custom-voice --speaker ryan -o output.wav
Model Details
- Architecture: Qwen3-TTS (Talker transformer + Code Predictor + Mimi speech tokenizer decoder)
- Parameters: 0.6B
- Quantization: None — bf16/fp16 weights (highest fidelity)
- Size: ~1.7 GB
- Sample rate: 24 kHz
- Codec rate: 12.5 Hz
- Voices: 9 preset speakers (
serena,vivian,uncle_fu,ryan,aiden,ono_anna,sohee,eric,dylan) + instruction-based style control
- Guide: soniqo.audio/guides/speak
- Docs: soniqo.audio
- GitHub: soniqo/speech-swift
- Library
- mlx
- Type
- qwen3_tts
- License
- apache-2.0
- Languages
- en, zh, ja, ko, fr, de
- Created
- Jun 21, 2026
- Updated
- Jun 21, 2026