Engineers at Snips, a French startup building private, on-device voice assistants, writing about their open-source NLU, hardware, and ML projects.
HN commenters mention Snips for its Rust-based tract library, its barometer-based underground location tracking, and its private-by-design voice assistant; one co-founder describes the Snips Air product.
SNIPS (FR) is looking for Data-scientists and Mobile Engineers (internships & fellowships possible)
#AI #ContextAwareness #NLP #PredictiveModelling #iOS #Android #IoT #Reactive
At Snips, our goal is to put an artificial intelligence in every device, and create beautiful interfaces that blend seamlessly into our daily lives.
To make this happen, we are assembling a team of passionate, creative people that strive to apply cutting edge research to real-world problems, building delightful products that also protect our privacy.
If you believe that AI can be beneficial, want to keep learning new technologies and actually make a difference with your work, then get in touch, we have something for you :-)
WHAT YOU GET
- A great place to work and experiment with your craziest ideas
- An opportunity to apply cutting-edge technologies to solve real world problems
- A team that is passionate, creative and at the top of their game
- Cool office in the center of Paris, with great amenities (cafeteria, gym, game room, ...)
- Above-market salary
- Stock options plan with additional grants each year
- Fully paid health insurance with the highest coverage
- Fully paid lunches in delicious eateries around the office
- Fully paid public transport card
- Any hardware you need to GTD
- Snips shoes and gears!
Curious? Check our job page for more details http://snips.net/jobs.html
- or check our blog: https://medium.com/snips-ai
- or jump straight into the process …
Maybe ten years ago, I read a blog post by, I think, a French company called snips, that explained how their app used the pressure sensor to detect when the train entered or leaved a station. It turned out there was a very clear signal due to the sudden pressure increase or decrease when the train entered or left the tunnel between the stations.
Edit: found it. https://medium.com/snips-ai/underground-location-tracking-3ea56803dddc
https://mycroft.ai/product/mark-ii/
Unfortunately it isn't going as well as they'd hoped:
https://mycroft.ai/blog/update-from-the-ceo-part-1/
Or you could try reSpeaker (different options available):
https://wiki.seeedstudio.com/ReSpeaker/
An older interesting comparison of mic arrays:
https://medium.com/snips-ai/benchmarking-microphone-arrays-respeaker-conexant-microsemi-acuedge-matrix-creator-minidsp-950de8876fda
And I do invite you to try out our voice platform, running 100% on-device on Raspberry Pi 3 or equivalent (and soon MCUs). We're on par with Amazon, Google et al. in terms of performance:
- https://makers.snips.ai/
- https://github.com/snipsco/snips-nlu
- https://medium.com/snips-ai/snips-nlu-is-an-open-source-private-by-design-alternative-to-dialogflow-amazon-lex-and-other-nlu-a95dbe16f4a1
Hi everyone!
I am a co-founder at Snips. This is a new product we are working on, which is due for next year
You can read more about it here https://medium.com/snips-ai/snips-air-a-private-by-design-open-source-decentralized-voice-assistant-a31e27fb799b
We will be publishing our whitepaper in the coming weeks, and are welcoming any feedback!
Cheers
This is the useful info I've seen hardware-wise recently:
https://medium.com/snips-ai/benchmarking-microphone-arrays-respeaker-conexant-microsemi-acuedge-matrix-creator-minidsp-950de8876fda
https://developer.amazon.com/alexa-voice-service/dev-kits/conexant-4-mic/
Amazon's 7-mic hardware has its own OEM program.