Voice-activity detection, speech to text, and text to speech all on RP2040

Article excerpt

Pete Warden is convinced local voice interfaces and sub-$1 embedded chips will fundamentally change how we interact with everything in the physical world. I’m so excited to introduce Moonshine Micro, a version of the Moonshine Voice open source framework that can run a useful voice interface in just 520KB of RAM. It contains separate libraries for voice-activity detection, speech to text, and text to speech, all powered by tiny neural networks with an example bringing them all together on an 80 cent Raspberry Pi RP2350 chip. This release runs a 50-word command recognizer, that’s fully trainable for custom words, and a neural network-based text to speech engine, and can be used to set up a wifi connection. See the video below and more in the article here.