# Generate audio with Stable Audio Open Small on LiteRT

## In this learning path

- [Introduction](https://learn.arm.com/learning-paths/mobile-graphics-and-gaming/run-stable-audio-open-small-with-lite-rt/)
- [Set up your development environment](https://learn.arm.com/learning-paths/mobile-graphics-and-gaming/run-stable-audio-open-small-with-lite-rt/1-prerequisites/)
- [Download and test the model](https://learn.arm.com/learning-paths/mobile-graphics-and-gaming/run-stable-audio-open-small-with-lite-rt/2-testing-model/)
- [Convert Stable Audio Open Small model to LiteRT](https://learn.arm.com/learning-paths/mobile-graphics-and-gaming/run-stable-audio-open-small-with-lite-rt/3-converting-model/)
- [Build LiteRT](https://learn.arm.com/learning-paths/mobile-graphics-and-gaming/run-stable-audio-open-small-with-lite-rt/4-building-litert/)
- [Create a simple program for Android target](https://learn.arm.com/learning-paths/mobile-graphics-and-gaming/run-stable-audio-open-small-with-lite-rt/5-creating-simple-program-for-android/)
- [Create a simple program for macOS target](https://learn.arm.com/learning-paths/mobile-graphics-and-gaming/run-stable-audio-open-small-with-lite-rt/6-creating-simple-program-for-macos/)
- [Next Steps](https://learn.arm.com/learning-paths/mobile-graphics-and-gaming/run-stable-audio-open-small-with-lite-rt/_next-steps/)

## About this Learning Path

| Skill level:    | Introductory       |
|------------------|-------------------|
| Reading time:    | 30 min            |
| Last updated:    | 21 Aug 2026       |

### Authors:
- Nina Drozd, Arm [GitHub](https://github.com/NinaARM) | [LinkedIn](https://linkedin.com/in/ninadrozd)
- Annie Tallund, Arm [GitHub](https://github.com/annietllnd) | [LinkedIn](https://linkedin.com/in/annietallund)
- Gian Marco Iodice, Arm
- Adnan AlSinan, Arm
- Aude Vuilliomenet, Arm

### Arm IP:
- [Cortex-A](https://support.arm.com/?tab=compute-ip&Product%20Type=Application%20Processors) 
- [Cortex-X](https://developer.arm.com/Processors#q=Cortex-X)

### Tags:
- [Performance and Architecture](https://learn.arm.com/tag/performance-and-architecture)
- [Linux](https://learn.arm.com/tag/linux)
- [Android](https://learn.arm.com/tag/android)
- [CPP](https://learn.arm.com/tag/cpp)
- [Python](https://learn.arm.com/tag/python)
- [Hugging Face](https://learn.arm.com/tag/hugging-face)

### Who is this for?
This is an introductory topic for developers looking to deploy the Stable Audio Open Small text-to-audio model using LiteRT on an Android™ device or on a reasonably modern platform with macOS®.

### What will you learn?
Upon completion of this Learning Path, you will be able to:
- Download and test the Stable Audio Open Small model.
- Convert the Stable Audio Open Small model to the LiteRT (.tflite) format.
- Compile the application for an Arm CPU.
- Create a simple application that generates audio.
- Run the application on an Android smartphone and generate an audio snippet.

### Prerequisites
Before starting, you will need the following:
- A Linux-based x86 or macOS development machine with at least 8 GB of RAM and 50 GB of disk space (tested on Ubuntu 22.04 with x86_64).
- A [HuggingFace](https://huggingface.co/) account.
- An Android phone in [developer mode](https://developer.android.com/studio/debug/dev-options) and a cable to connect it to your development machine.

### Summary
You’ll convert Stable Audio Open Small to LiteRT (`.tflite`) and run its audio-generation application on Android or macOS Arm devices. First, you’ll prepare the workspace, download the model files, and convert its Conditioners, DiT, and AutoEncoder submodules. Then, you’ll build LiteRT and FlatBuffers, compile the application for your chosen platform, run it with a text prompt, and retrieve the generated `output.wav` file.

### Frequently asked questions

#### How do I know the model download is complete before conversion?
Verify that `model_config.json` and `model.ckpt` are in your workspace directory. Run the listed `ls` command before you start conversion.

#### Where do I run CMake to build the Android app?
From your workspace, navigate to `ML-examples/kleidiai-examples/audiogen/app` and create a build directory. Run CMake from that build directory.

#### Which Android ABI should I use for Arm targets?
Set `ANDROID_ABI=arm64-v8a` in the CMake command for the Android build.

#### What paths do I pass to CMake after building LiteRT?
Build LiteRT and FlatBuffers, then configure CMake with `TF_INCLUDE_PATH`, `TF_LIB_PATH`, and `FLATBUFFER_INCLUDE_PATH`.

#### What result should I expect when running the program on the device?
Run `audiogen` on the device with a text prompt. After inference, pull `/data/local/tmp/app/output.wav` to your host machine.