Piper
Voice & transcriptionA fast, local open-source text-to-speech system designed to run well on small devices.
Piper is a lightweight neural text-to-speech system built for speed and low resource use — it runs well even on a Raspberry Pi. It ships voices across many languages and is a common choice in offline voice assistants.
It trades the last degree of naturalness for being genuinely fast, small, and fully local.
Where it's ideally used
A fit for offline or embedded voice — devices, kiosks, local assistants — where speed and footprint outrank polish.
Where it doesn't fit
Not the choice when voice quality is front and centre of the user experience.