Air voice and senses
- Live calls. A hands-free voice and webcam call. Whisper hears you, the reply is spoken as it streams about 3 seconds after you stop, and you can talk over it.
- Expressive voices. Kokoro and VoxCPM2 voices, or clone one from a 5 to 30 second recording.
- Webcam vision. Ask about what the camera sees, with live captions.
- Transcribe. Drop in a recording and get every word, who said it, and a summary with action items.
- Translate. Any of 150 languages, including text inside a picture.











