This is a series of articles looking at the ASRock Industrial NUC BOX-255H running Linux. In this series I examine every aspect of this Mini PC in detail from a Linux perspective.
One of the NUC’s aspects that makes it interesting from a hardware perspective is its GPU and NPU capabilities. The machine has the Intel Arc 140T, the iGPU used in the Intel Arrow Lake H/HX processor series. It’s quite a powerful integrated graphics setup. It shares the system DDR5 memory, it has 1024 cores, 64 TMUs, and 32 ROPs. It also sports 128 tensor cores, and 8 ray tracing cores.
This article looks at the OpenVINO AI Plugins for Audacity, a popular open-source audio editor and recorder. The AI plugins provide a set of AI-enabled effects, generators and analyzers. The plugins run solely on the NUC Box-255H with no internet connection needed.
OpenVINO (Open Visual Inference and Neural Network Optimization) runs the AI models on the Arc 140T. OpenVINO is an open-source software toolkit developed by Intel that optimizes and deploys deep learning models.
The plugins provided the following functionality:
- Music Separation — Separate a mono or stereo track into individual stems — Drums, Bass, Vocals, and Other Instruments.
- Noise Suppression — Removes background noise from an audio sample.
- Music Generation and Continuation — Generates music from a text prompt or continues an existing song.
- Whisper Transcription — Uses whisper.cpp to generate a label track containing the transcription or translation for a given selection of spoken audio or vocals. Whisper is an awesome automatic speech recognition (ASR) system trained on many hundred of thousands of hours of multilingual and multitask supervised data collected from the web. Powered by deep learning and neural networks, Whisper is a natural language processing system that’s built on PyTorch.
- Super Resolution – another interesting plugin which upscales and enriches audio for improved clarity and detail.
Installation
In my previous article I noted that getting Stable Diffusion working with the Arc 140T iGPU took a bit of research. That’s not the case with the AI Plugins for Audacity. It’s a very simple installation with the Ubuntu distribution. Install Audacity using snap.
$ sudo snap install audacity
I want to experiment with NPU support so, let’s install the intel-npu-driver tool.
$ sudo snap install intel-npu-driver
The AI models are not included in the audacity snap. But they can all be installed with this command.
$ sudo audacity.fetch-models --batch
Your user account needs to be a member of the render group. If you don’t see render when you issue the command groups, follow the project’s GitHub’s README page. I’m already a member of the render group, as confirmed below.
Three of the AI features are accessed from the menu Effect / OpenVINO AI Effects. Music Generation is accessed from Generate. And the OpenVINO Whisper Transcription is available in Analyze.
I’m going to take a look at a few of the plugins. Let’s start with music separation.
Next page: Page 2 – Music Separation
Pages in this article:
Page 1 – Introduction and Installation
Page 2 – Music Separation
Page 3 – Music Generation and Summary
Complete list of articles in this series:
| ASRock Industrial NUC BOX-255H | |
|---|---|
| Introduction | Introduction to the series and interrogation of the NUC BOX-255H |
| Benchmarks | Benchmarking the NUC BOX-255H |
| Power | Testing and comparing the power consumption under various workloads |
| Stable Diffusion | Deep Learning with Stable Diffusion |
| Audacity AI Plugins | Let's explore OpenVINO AI Plugins for Audacity |