Voice Recognition and Speech Synthesis on an RPi Pico 2

Per Adafruit Blog, Pete Warden is pursuing an ambitious goal: running voice recognition and text-to-speech on a fifty-cent chip. A recent project brings him closer, demonstrating both functions working entirely on a Raspberry Pi Pico 2—a microcontroller with just over half a megabyte of RAM and four megabytes of flash. Most voice assistants need WiFi, cloud APIs, and powerful processors, but this implementation handles both duties on the Pico 2, using Adafruit microphone and amplifier breakouts. It shows that voice interfaces are possible on extremely constrained hardware, without cloud infrastructure.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton