ModelListedExternal

Qwen3 Tts Custom Voice

A fine-tuned version of Qwen3-TTS-12Hz-1.7B-Base trained on a custom speaker dataset using supervised fine-tuning (SFT).

Sourcehuggingface.co/thunk6/qwen3-tts-custom-voice/tree/3c369a01d6603fe3b031b5f6bf9e13eff33898e3

KindText modelBase modelQwen/Qwen3-TTS-12Hz-1.7B-BaseVersion

Model source

Kind: Text model
File size: 3.57 GB
Base model: Qwen/Qwen3-TTS-12Hz-1.7B-Base
Version: vmodel
Updated: Sep 13, 2026
Downloads: 11
Source: Hugging Face

Source description

Qwen3-TTS — Custom Voice Clone

A fine-tuned version of Qwen3-TTS-12Hz-1.7B-Base trained on a custom speaker dataset using supervised fine-tuning (SFT).

Requirements

pip install qwen-tts torch torchaudio

Usage

import torch
import soundfile as sf
from qwen_tts import Qwen3TTSModel

tts = Qwen3TTSModel.from_pretrained(
    "thunk6/qwen3-tts-custom-voice",
    device_map="cuda:0",
    dtype=torch.bfloat16,
)

wavs, sr = tts.generate_custom_voice(
    text="Hello, this is my cloned voice.",
    speaker="my_custom_voice",
)

sf.write("output.wav", wavs[0], sr)

Training Details

Setting	Value
Base model	Qwen3-TTS-12Hz-1.7B-Base
Fine-tuning method	SFT
Learning rate	2e-6
Batch size	4
Epochs	10
Hardware	A100 80GB

Sources

1 source

Hugging FacePrimaryVersion 3c369a01d6603fe3b031b5f6bf9e13eff33898e3

huggingface.co/thunk6/qwen3-tts-custom-voice/tree/3c369a01d6603fe3b031b5f6bf9e13eff33898e3

Verified Sep 13

Model artifacts

1 artifact

model.safetensors

safetensors · 3.57 GB · SHA-256 1c3e93022b4b…1e84 · Hugging Face

Download file

Source excerpts

2 excerpts

ReadmeSep 13huggingface.co/thunk6/qwen3-tts-custom-voice

--- license: apache-2.0 base_model: Qwen/Qwen3-TTS-12Hz-1.7B-Base tags: - text-to-speech - qwen3-tts - fine-tuned - voice-cloning language: - en --- # Qwen3-TTS — Custom Voice Clone A fine-tuned version of [Qwen3-TTS-12Hz-1.7B-Base](https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-Base) trained on a custom speaker dataset using supervised fine-tuning (SFT). --- ## Requirements ```bash pip install qwen-tts torch torchaudio ``` ## Usage ```python import torch import soundfile as sf from qwen_tts import Qwen3TTSModel tts = Qwen3TTSModel.from_pretrained( "thunk6/qwen3-tts-custom-voice", device_map="cuda:0", dtype=torch.bfloat16, ) wavs, sr = tts.generate_custom_voice( text="Hello, this is my cloned voice.", speaker="my_custom_voice", ) sf.write("output.wav", wavs[0], sr) ``` ## Training Details | Setting | Value | |---|---| | Base model | Qwen3-TTS-12Hz-1.7B-Base | | Fine-tuning method | SFT | | Learning rate | 2e-6 | | Batch size | 4 | | Epochs | 10 | | Hardware | A100 80GB |

NoteSep 13huggingface.co/thunk6/qwen3-tts-custom-voice

Source context: 11 downloads · 0 likes · Pipeline text-to-speech · Repo thunk6/qwen3-tts-custom-voice