@robjama Thanks brother! The Snips Voice Platform allows anyone to build an AI powered voice assistant with ease. This is the first end-to-end platform to offer on-device ASR and NLU, while protecting user privacy. The Snips Assistant can be configured easily through our web console, which includes the ability to select from pre-built assistants or to create completely custom ones. The assistant can then be deployed to a RaspberryPi, to Linux or Android, with more platforms coming soon. Come give it a try! www.snips.ai

Check it out in action here:

@chevreuil totally, it's time we stop saying we have to chose between AI and Privacy!

@randhindi @chevreuil I would never say we don't need more privacy, but I wouldn't say this is the first AI-powered voice assistant "with privacy". That's misleading. Siri already protects your privacy extremely well. For instance, "When we do send information to a server, we protect your privacy by using anonymized rotating identifiers so that searches and locations can’t be traced to you personally." https://www.apple.com/privacy/ap...

@benjmerritt thanks 🙌 There are two ways to use Snips today: - as a maker, or for prototyping on a Raspberry Pi, you can use our web console and build your assistant yourself, for free - as a commercial product maker, we can make a build that will run on your device, on any platform (Android, Linux, etc..) The whole thing is written in Rust, so it can cross-compile and works super fast!

@scottbowler We are building the platform itself, in order for companies to become independant of Google, Amazon, Apple, Microsoft. The idea is to provide the same or better quality of AI while respecting user privacy. We will try to be compatible with the Alexa intents to allow device builders to switch easily to our platform

@mael_p Still just getting my head around this (sorry!) - so I have a platform that allows customers to manage content. Could I use Snips to allow my customers to control their actions via voice control, or does each customer need to use snips and integrate it with my platform?

@scottbowler Our platform is made for connected devices, so a device maker would use Snips to translate speech into a structured meaning representation which allows his software to understand the meaning of the user query. He can then do what he wants with the query, and can for instance integrate to your platform. We will later build a Skill Store which allows service providers like you to provide pre-built code that device makers can add in 1 click to their device if they want to let their users integrate easily with your platform

@mael_p Sounds a very interesting concept - I'll definitely be keeping a close eye on it - good luck with the launch!

@rahafharfoush Thanks so much!! The team is ecstatic seeing all of your support!

@javimaker @rahafharfoush Thanks Rahaf xx

@laurent_blrs That's amazing to hear, thanks Laurent!

@badihsaikali thanks Badih 🙌

@bourgois merci!!

@danr_4 Hey there Dan, the underlying technology has been developed from the app into our new Voice Assistant Platform. After the launch of the app, we had an overwhelming number of companies inquire how they could include our tech in their products, so rather than pursue the mobile app route, we launched the platform. Now makers, prototypers, and enterprises can integrate a voice assistant into their products with ease.

@tom_downs95 Thanks Tom! We think so too, and we believe privacy is a key trait of it as well!

@joniofir thanks :)

@jhamilius Thanks Julien!! :)

@lilly_cally Thanks!

@gflandre we actually have an auto-tagging feature that leverages the self learning abilities of our engine. We will also soon add a killer feature that enables you to automatically augment your dataset to get better performances out of the box! For the resilience, it turns out to be important for devices such as cars and home appliances that must respond at all time

@gflandre We will add ways for users who accept to share data to give queries and improve their NLU too. Many devices might not be connected to the net at all time, think cars, or when you are travelling in a remote area. But first and foremost we don't want the variable costs of cloud subscriptions and we want to keep data safe

@randhindi @mael_p That's awesome if in the future the NLU can be improved! Indeed I didn't have the "cars" use-case in mind but that makes a ton of sense. Of course in terms of privacy/performance/cost, this is great!