We are building local speech-to-speech pipelines to benefit humanity. We believe that voice-first communications are the next step in computer usage and we do our best to make sure that you will never need to touch a keyboard again.

Blog

Read our latest news

  1. Your ASR already knows when you're done talking
  2. Running Voxtral Realtime on a single consumer GPU
  3. Measuring turn-taking latency without a stopwatch
enru