February 10, 2026
The NVIDIA DGX Spark is a desktop GPU workstation with 128GB of unified memory.
Working with the team at @NVIDIAAIDev, we've been using these little powerhouse machines for voice agent development, testing new models and inference stacks, and training LLMs and audio models.
Today we published a guide to training the Smart Turn model on the DGX Spark. Smart Turn is a fully open source (and open training data), native audio turn detection model that supports 23 languages.
The guide walks you through installing the right dependencies for this new Arm + Blackwell architecture, and includes benchmarks comparing training times on various platforms.
If you're interested in a training audio models and small LLMs, the post is a nice resource.
Blog post link[1]
The Smart Turn open source turn detection model[2]