ashwmurt/neutts_nano

NeuTTS-Nano: On-device text-to-speech language model with instant voice cloning

NeuTTS-Nano is a compact on-device text-to-speech language model from Neuphonic. It pairs a small Llama-family causal LM backbone with the NeuCodec 50Hz neural audio codec, and produces natural-sounding 24kHz speech with instant voice cloning from a short reference clip. The backbone exports as two graphs (a 128-token prefill and a single-token decode) linked into one weight-shared context binary; phonemization, prompt assembly, sampling and NeuCodec encode/decode run on CPU in the app.

This is based on the implementation of NeuTTS-Nano found here. This is a standalone recipe compatible with the Qualcomm® AI Hub Models CLI — it can be installed, compiled, and evaluated on real Snapdragon devices via Qualcomm® AI Hub Workbench.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Setup

1. Install the package

Install the base package, fetch this recipe from Hugging Face, then use the qai-hub-models CLI to install the recipe's dependencies:

# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
pip install qai-hub-models
qai-hub-models register ashwmurt/neutts_nano
qai-hub-models install neutts_nano

register downloads the recipe and names it neutts_nano, which is how every command below refers to it.

2. Configure Qualcomm® AI Hub Workbench

Sign-in to Qualcomm® AI Hub Workbench with your Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.

With this API token, you can configure your client to run models on the cloud hosted devices.

qai-hub configure --api_token API_TOKEN

Navigate to docs for more information.

Run CLI Demo

Run the following simple CLI demo to verify the model is working end to end:

qai-hub-models demo neutts_nano

More details on the CLI tool can be found with the --help option. See demo.py for sample usage of the model including pre/post processing scripts.

Export for on-device deployment

To run the model on Qualcomm® devices, you must export the model for use with an edge runtime such as TensorFlow Lite, ONNX Runtime, or Qualcomm AI Engine Direct. Use the following command to export the model:

qai-hub-models export neutts_nano --target-runtime qnn_context_binary --precision float --device "Samsung Galaxy S25 (Family)"

Additional options are documented with the --help option.

License

• The license for the original implementation of NeuTTS-Nano can be found here.

References

• NeuCodec Neural Audio Codec for Speech Language Models
• Source Model Implementation

Community

• Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
• For questions or feedback please reach out to us.