EMKAY
A stock OpenVoiceOS voice assistant, running in your browser. Say "Hey Mycroft" – on your phone, tablet, TV or a Raspberry Pi.
Early alpha · free and open source · nothing to install
What it is
Emkay is OpenVoiceOS – the open, privacy-minded voice assistant – running entirely inside a web page, with the Mycroft Mark II face as its screen. A phone, a tablet, a TV or a Raspberry Pi in kiosk mode becomes an assistant without an app store, sideloading or setup on the command line.
"Emkay" is "MK" said out loud – a nod to Mycroft's Mark I and Mark II.
Get started
- Open app.emkay.ai. The first start downloads OpenVoiceOS and takes a few minutes; later starts are quicker.
- Tap the screen when it says so – browsers only allow sound and the microphone after a tap.
- Choose Set up to pick language, voice and more, or start with the defaults (your phone's language and OpenVoiceOS' public speech servers).
- Say "Hey Mycroft", tap the face, or tap ⌨ to type. Everything else is under ⚙.
- ⚙ → Install as app puts Emkay on your home screen (on an iPhone: Share → Add to Home Screen).
What works today
OpenVoiceOS, unchanged
OVOS core with the installer's default skills, the homescreen and the Mark II face, from the alpha channel.
"Hey Mycroft"
A wake word that runs in the browser, so it listens without a tap.
Skills that ask back
"Set an alarm" – "for when?" – "in five minutes". Follow-up questions work, as on a device.
The control panel
The stock OVOS control panel: settings, setup wizard, backups, installing skills and plugins.
A skill store
store.emkay.ai: OVOS skills for Emkay, by category, in your language, installed with one tap.
The phone as hardware
Volume, location, power, sensors (battery, network, screen, motion, light) and the camera.
Languages: English, Danish, German, French, Italian, Spanish, Galician, Catalan, Basque, Portuguese (Portugal and Brazil) and Dutch.
Privacy
OpenVoiceOS runs in your browser, on your device. With the default setup, what you say is sent to OpenVoiceOS' public speech-to-text and text-to-speech servers; the control panel lets you choose others. Settings live in the browser's own storage on the device.
Coming
- Speech recognition and voices that run on the device itself.
- A store that only offers skills tested to work in Emkay, also in the control panel.
- HiveMind: a phone or tablet in every room as a satellite of one OpenVoiceOS in the house.