StudioFlow CORE is our smart assistant reference platform, built to explore what’s possible with a voice-enabled assistant that doesn’t rely on the tech giants of Amazon, Google, Apple or Microsoft.
If you’re interested in building a voice experience, but would prefer to own your own data, have strict information security needs or just feel like your business is best served owning as much of the tech stack as possible, then CORE could be for you.
Built on the PiCroft platform from open source smart assistant project Mycroft, CORE is a set of customisations and configurations that allow experimentation on what’s possible so much quicker.
CORE is an experimentation platform. Much of our work designing for voice involves prototyping potential unique solutions to unmet user needs. There just aren’t the prototyping platforms out there to do this kind of high-fidelity concept evaluation – so we built CORE to address these needs.
- Custom hardware (e.g. bespoke smart speakers)
- Linux: Ubuntu/Debian, Arch, Fedora
- Supports custom voices (either off-the-shelf or bespoke using our ‘YourVoice’ product)
- Locked-down set of intents (constraining the domain, improving accuracy and ensuring a tight good-quality UX)
- Custom wake word (maybe even the option to give each user their own wake word – the name of a loved one, or favourite celebrity!?)
- Options to personalise content for the persona of user (news, weather, TV listings, etc…)
- Cellular data connection, for users or locations where in-home broadband isn’t available.
- Relatively cheap hardware (we’re looking to get the cost of these devices down to around £100-£150 + yearly maintenance)
- Options to add additional sensors: motion detection, fall detection, raised voices, push-to-wake, IFTTT integration, etc…
If you’re interested in having a conversation with us about what’s possible with CORE, and how we can help you explore embedding voice into your products, get in touch: