uralstech/Qwen3-TTS-12Hz-0.6B-Base-litert-lm-omni
Qwen3-TTS-12Hz-0.6B-Base for LiteRT-LM Omni
This repository contains the Qwen3-TTS-12Hz-0.6B-Base model converted to the format required by the LiteRT-LM Omni TTS Engine.
The model is intended for on-device text-to-speech (TTS) inference using LiteRT-LM Omni.
Model
This model is converted from:
The original Qwen3 TTS model provides multilingual text-to-speech capabilities. This converted version packages the model components in the format expected by the LiteRT-LM Omni TTS Engine.
Files
The repository contains all files required by the LiteRT-LM Omni TTS Engine.
The model components include the LiteRT/TFLite models and supporting data required for text processing, speech generation, and codec decoding.
All files in this repository are required for inference. Do not remove individual files when deploying the model.
Usage
This model is designed to be used with the LiteRT-LM Omni TTS Engine rather than directly with the original Qwen3 TTS Python implementation.
Conversion
The model was converted from the original LiteRT Qwen3 TTS model using the conversion scripts here.
The conversion process converts the original model artifacts into the directory structure and file formats expected by the LiteRT-LM Omni TTS Engine.
License
This model is released under the Apache 2.0 license, inherited from the base model.
Credits
- Qwen3 TTS: Qwen
- LiteRT model: litert-community/Qwen3-TTS-12Hz-0.6B-Base
- Runtime: LiteRT-LM Omni
Disclaimer
This repository contains a converted version of the upstream model. For information about the model architecture, training, intended use, limitations, and responsible use, please refer to the upstream model card.