Show HN: Live speech-to-text on an ESP32-P4 microcontroller, no cloud
Developers have demonstrated a live speech-to-text system that runs entirely on an ESP32-P4 microcontroller, eliminating the need for cloud processing. The solution uses on-device neural inference to convert spoken words into text in real time, leveraging the chip’s AI acceleration capabilities. The project’s source code is publicly available, allowing others to replicate or extend the offline transcription capability.