WHOOP does not have the presence in the wearable space as other brands, but in certain circles, it’s a household name. Their business model requires you to have a yearly app subscription to use their fitness tracker, but here at Hackaday, we are big fans of actually owning the devices you buy — which is why we were happy to hear about an open source and subscription free WHOOP compatible app!

The goal of the so-called OpenStrap project is not to re-create the WHOOP app. Rather, the algorithms and processing methods are developed from scratch, based on public research. It’s all calculated locally on a 1 Hz interval, based on the data the WHOOP 4.0 device feeds the app. As such, the health data collected from the watch, never leaves the phone. While not the main goal of the project, the privacy improvement of the app’s serverless nature cannot be overstated. However, to display metrics, you first need to get data off the WHOOP to begin with.

The crux of the issue with making the WHOOP 4.0 work without the official app is the reliance on proprietary Bluetooth protocols. Fortunately, the protocol itself ended up being relatively simple. The WHOOP 4.0 amounts to little more than a series of sensors that sit on the user’s wrist. As such, the app can subscribe to the Bluetooth feed and decode the data, right? Well, the devil is always in the details with such things, and the protocol came with its fair share of quirks. The hardware clock needs to be synchronized, or it simply defaults to zero Unix time. Moreover, the analog sensors like, ambient temperature are given in relative ADC values, and are not terribly useful without calibration. Regardless, the result of the reverse engineering effort speaks for itself with the OpenStrap app able to recreate much of the functionality in WHOOP’s official app.

Quite often, devices reliant on proprietary apps are little more than manufactured e-waste. While we don’t expect many of you to actually own a WHOOP 4.0, we do hope to see the OpenStrap project keep at least a few out of the landfill in the future.

Speaking is much faster than typing, and while it’s an increasingly convenient way to interact with computers, it’s hardly private. Providing speech privacy in a way we haven’t seen before is this prototype tongue-reading system that uses machine learning and ultrasound to read tongue movements and turn them into decoded speech. Not only can a user speak without emitting a sound, since it doesn’t read sound waves it’s completely immune to noisy environments.

It turns out that tongue movements are a very rich source of information about speech, and an ultrasound probe under the chin takes very clear video of a tongue. With a dataset consisting of only around 50 hours of training data, the system has a 15.6% error rate and generalizes across different speakers (as long as they speak with similar accents).

That error rate may seem high at first glance, but keep in mind this is for a prototype system built in a month around a relatively small training dataset. All indications are that better results are just a matter of better training.

Probably the biggest drawback at the moment is the size of the ultrasound probe and the way it must be held under one’s chin like a contact microphone, but at the moment the probe is an off-the-shelf model that is hardly optimized for either size, weight, or wearability. If the system seems promising enough, a probe resembling an adhesive patch might even be possible.

It’s certainly a different approach from others we’ve seen in the past, including whispering while inhaling and reading lip and mouth movements.