#STT

Tagged: STT

Engineering

We benchmarked 13 speech-to-text APIs across 12 languages — and published everything

A preregistered, vendor-funded-but-verifiable STT benchmark: frozen methodology, 28,400 raw transcriptions, all scoring code public. Including the languages where our own engine loses.

Read more →
Essay

Are you using voice input? — On the next input interface after the keyboard

Caring about voice input the way we care about our keyboards. An essay tracing the journey from discovering AquaVoice to building koedesk, and through a parade of STT models.

Read more →