Qwen3-TTS

Summary

• Introduction
• Model Architecture
• Released Models
• Parameters
• Usage for developers
• Reference
• License
• Citation

Introduction

Qwen3-TTS covers 10 major languages (Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, and Italian) as well as multiple dialectal voice profiles to meet global application needs. In addition, the models feature strong contextual understanding, enabling adaptive control of tone, speaking rate, and emotional expression based on instructions and text semantics, and they show markedly improved robustness to noisy input text. Key features:

• Powerful Speech Representation: Powered by the self-developed Qwen3-TTS-Tokenizer-12Hz, it achieves efficient acoustic compression and high-dimensional semantic modeling of speech signals. It fully preserves paralinguistic information and acoustic environmental features, enabling high-speed, high-fidelity speech reconstruction through a lightweight non-DiT architecture.
• Universal End-to-End Architecture: Utilizing a discrete multi-codebook LM architecture, it realizes full-information end-to-end speech modeling. This completely bypasses the information bottlenecks and cascading errors inherent in traditional LM+DiT schemes, significantly enhancing the model’s versatility, generation efficiency, and performance ceiling.
• Extreme Low-Latency Streaming Generation: Based on the innovative Dual-Track hybrid streaming generation architecture, a single model supports both streaming and non-streaming generation. It can output the first audio packet immediately after a single character is input, with end-to-end synthesis latency as low as 97ms, meeting the rigorous demands of real-time interactive scenarios.
• Intelligent Text Understanding and Voice Control: Supports speech generation driven by natural language instructions, allowing for flexible control over multi-dimensional acoustic attributes such as timbre, emotion, and prosody. By deeply integrating text semantic understanding, the model adaptively adjusts tone, rhythm, and emotional expression, achieving lifelike “what you imagine is what you hear” output.

Model Architecture

Released Models Description and Download

Below is an introduction and download information for the Qwen3-TTS models that have already been released. Other models mentioned in the technical report will be released in the near future. Please select and download the model that fits your needs.

Tokenizer Name	Description
Qwen3-TTS-Tokenizer-12Hz	The Qwen3-TTS-Tokenizer-12Hz model which can encode the input speech into codes and decode them back into speech.

Model	Features	Language Support	Streaming	Instruction Control
Qwen3-TTS-12Hz-1.7B-VoiceDesign	Performs voice design based on user-provided descriptions.	Multilingual (10 languages)	✅	✅
Qwen3-TTS-12Hz-1.7B-CustomVoice	Provides style control over target timbres via user instructions; supports 9 premium timbres covering various combinations of gender, age, language, and dialect.	Multilingual (10 languages)	✅	✅
Qwen3-TTS-12Hz-1.7B-Base	Base model capable of 3-second rapid voice clone from user audio input; can be used for fine-tuning (FT) other models.	Multilingual (10 languages)	✅
Qwen3-TTS-12Hz-0.6B-CustomVoice	Supports 9 premium timbres covering various combinations of gender, age, language, and dialect.	Multilingual (10 languages)	✅
Qwen3-TTS-12Hz-0.6B-Base	Base model capable of 3-second rapid voice clone from user audio input; can be used for fine-tuning (FT) other models.	Multilingual (10 languages)	✅

Parameters

Inputs

• prompt - (text): The text content that will be converted to speech.
• language - (text): The language used to synthesize the speech. Supported values: Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, Italian, ...).
• ref_audio - (text): A reference audio file used for voice cloning. The model extracts voice characteristics (tone, pitch, speaking style) from this audio. Required for Base models; not used by VoiceDesign or CustomVoice variants.
• ref_text - (text): The transcript corresponding to the ref_audio. This helps the model better align the voice style with the spoken content.

Output

• output - (audio -.wav): The generated speech audio file synthesized from the prompt, using the voice characteristics extracted from ref_audio.

Usage for developers

Please find below the details to track the information and access the code for processing the model on our platform.

Requirements

pip install -r requirements.txt

Code based on AIOZ structure

import torch
import soundfile as sf
from qwen_tts import Qwen3TTSModel

def do_ai_task(
        prompt: Union[str, str],
        language: Union[str, str],
        ref_audio: Union[str, str],
        ref_text: Union[str, str],
        model_storage_directory: Union[str, Path],
        device: Literal["cpu", "cuda", "gpu"] = "cpu",
        *args, **kwargs) -> Any:
    """Define AI task: load model, pre-process, post-process, etc ..."""
    # Define AI task workflow. Below is an example
    device = "cuda" if torch.cuda.is_available() else "cpu"
    dtype = torch.float16 if device == "cuda" else torch.float32

    model_weights = model_storage_directory + "..."
    model = Qwen3TTSModel.from_pretrained(
        model_weights,
        device_map=device,
        dtype=dtype,
        attn_implementation="flash_attention_2",
    )

    wavs, sr = model.generate_voice_clone(
        text=prompt,
        language=language,
        ref_audio=ref_audio,
        ref_text=ref_text,
    )

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

    output_file = open("output.wav", "rb")  # io.BufferedReader
    return output_file

Reference

This repository is based on and inspired by Qwen's work. We sincerely appreciate their generosity in sharing the code.

License

We respect and comply with the terms of the author's license cited in the Reference section.

Citation

@article{Qwen3-TTS,
  title={Qwen3-TTS Technical Report},
  author={Hangrui Hu and Xinfa Zhu and Ting He and Dake Guo and Bin Zhang and Xiong Wang and Zhifang Guo and Ziyue Jiang and Hongkun Hao and Zishan Guo and Xinyu Zhang and Pei Zhang and Baosong Yang and Jin Xu and Jingren Zhou and Junyang Lin},
  journal={arXiv preprint arXiv:2601.15621},
  year={2026}
}