This demonstration by MosChip and Cadence shows a Small Language Model (SLM) voice assistant running entirely on-device on the Cadence Tensilica Vision Q7 DSP within an Axera AX650N platform – with no cloud connection. It walks through the full interaction loop: spoken input is converted to text, a compact quantized language model (SLM) generates a reply, and text-to-speech voices the response back to the user. The demo highlights how conversational AI can be added to camera-bearing edge products by reusing the Cadence Vision DSP already on the chip, instead of adding a dedicated accelerator or depending on the cloud. The result is responsive, private, power-efficient voice interaction that keeps data on the device.
