Qwen / VOICE / TTS

Qwen3-TTS-12Hz-0.6B-Base

The small Qwen3-TTS Base checkpoint supports multilingual speech synthesis from reference voice audio.

Source reviewed 2026-10-24 · AiBrainWorX editorial summary

Creator: Qwen
Parameters: 0.6B
Context: Task-specific; see the reviewed source and runtime
License: apache-2.0

01 / CAPABILITIES

What it is built for

Local speech synthesis
External task-specific runtime; not integrated into launcher chat

Inputs: Text → audio

Runtime limits and on-device memory must be validated separately.

02 / ON YOUR MACHINE

Very small model / edge candidate; runtime validation required

Uses an audio tokenizer and dedicated Qwen TTS runtime. The official Base-model training recipe is single-speaker, separate from text LoRA.

Launcher integration

External task-specific runtime; not integrated into launcher chat

Runtime and training workflow

qwen-tts

External Qwen TTS single-speaker SFT. Training suitability requires exact checkpoint format and prepared data; a catalog entry is not a completed training test.

Compare dataset preparation guides →

03 / WHAT IS VERIFIED

Evidence, with context

Our test status

Source reviewed only. No local model or Android inference test performed.

We have not added a benchmark number without documenting its test conditions. A model-card capability is not a measured speed or quality guarantee on your hardware.

04 / OWNERSHIP & USE

apache-2.0

Review the exact checkpoint, voice assets and applicable upstream notices; model references do not grant new rights.

AiBrainWorX authors this profile and the launcher. Qwen remains the model’s creator. Models are not rebranded as our own weights.

05 / GO DEEPER

Original sources & model files

This page is our concise summary, not a mirror of the full upstream card. Review the original documentation and exact artifact before downloading.

Official card on Hugging Face ↗Files & versions ↗Reviewed card revision ↗

Reviewed revision: 5d83992436ea · Specifications may change after the review date.