Smart appliance manufacturer Wyze makes very affordable security cameras that do the basics just fine, but have lacked the features bigger names have been promoting on similar products with bigger price tags. But that may be about to change as the company is testing out AI-powered vehicle and people recognition for the Wyze Cam v2 and Wyze Cam Pan.
Google Fit is something Google has been working diligently on since before its announcement at Google I/O. The service itself doesn't seem to be very well-rounded yet, but with a preview SDK available to developers for a few months and an app in the works, it seems like things are headed in the right direction.
We've seen only glimpses of the Fit app so far from leaked product reviews and comparison videos, but now that the whole app has been made available in a leaked Nexus 6 dump that's floating around (we're not going to distribute test-keys signed APKs), we can finally take a quick look at the app itself.
We've all played the "Wait, who is that guy again?" or "What song is that?" while watching a movie. Now, with the latest update to the Play Movies app, if you want to get the answer to that question, all you have to do is press pause. Info cards will then pop up with face recognition of actors on screen, what other movies they're in, and what music you're hearing.
The feature only works on "supported movies" and it's a little unclear which ones those are (there doesn't seem to be any icon or indicator that shows whether any given movie in your library can show cards).
Did you know that, since the last update to Google Search, developers have been able to utilize offline voice recognition? Previously, any non-system app that wasn't an IME (Input Method Editor) that hoped to recognize your voice without a web connection needed a rather kludgy typing overlay. Since the update though, apps can hear and interpret not just your words, but essentially any command that doesn't explicitly require web access.
Utter!, taking advantage of the new possibility, claims to have become the first app with working offline voice recognition for Android Jelly Bean. After reaching out to Ben Randall, the app's developer, we've got a clearer picture of what it used to take to recognize speech offline, and what has changed.
Google Translate, the frequently-overlooked wonder app of the 3rd millennium, got some new features today. Chief among them is an amazing new image-based translation mechanism. The app now supports use of your camera to take a picture of the text you would like to translate. Once that's done, just "brush" over the word or phrase you need to read and Translate will do what it says on the tin: render that text in your preferred language. It's not quite WordLens' live view, but then, if you've ever used WordLens' pause feature, you know that slowing things down a bit might actually be helpful.
Thumb Keyboard, one of the most intuitive, well-designed, and practical keyboards available (especially for tablet users) got a big update recently, bringing the app up to version 4.5.
Perhaps the biggest benefit to the new update is that Thumb Keyboard now supports ICS' continuous voice recognition, which in case you've forgotten, is the feature introduced with Ice Cream Sandwich that actively listens and dictates your speech. This is a feature I haven't seen in other alternative keyboards, and it's definitely nice to have. Other big additions include gesture support as well as a new one-handed layout for phones with larger displays.
In its endless attempts to make searching easier for everyone, Google has introduced yet another way to search via its mobile site at google.com: handwriting recognition. If you go to Google's search page from your phone or tablet's mobile browser and enable the feature via settings, you can now scribble your searches on the screen, even after receiving results. It's pretty fancy!
Of course, this does raise the question of whether this input method is any faster. In the video above, in an attempt to show how this might be used, we see a man who has had nearly all of his fingers broken scribble the words "ski lessons" on the screen.
Amazon-owned development house A9 Innovations has released a product search app built on the idea that instead of tapping buttons to take pictures of products, you'd rather just point your camera at products. Probably not a bad notion! Not exactly the most important thing to spend a bunch of money and time developing, but hey, if you can just wave your phone in front of a movie and get pricing and review information, it's gotta be worth it, right?
Well, it would be, if it worked like that. Unfortunately, in practice the augmented reality app has some trouble recognizing cover art.
Want a add a calendar appointment? Tell Utter, and it'll take care of it. Get travel details, find out the weather, and launch applications - all child's play for Utter, and all done using native applications instead of just simple searches.
Actually, that's just scratching the surface for Utter - if you watch the entire 22 minute video, you'll see the developer, brandall of XDA, use Utter to set his CPU governor (yes, it asks for superuser permissions), reboot into the bootloader, and even have a conversation.
OK, this is probably one of the cooler apps I've seen in a while. ObscuraCam is an open source photo anonymizer project made in concert with the folks over at Witness.org - a human rights violation watchdog. Why is a human rights organization helping you blur those photos you took that night you accidentally [insert potential human rights violation here] but still thought would be hilarious to upload to Facebook?
Well, Witness wants this app to be used by political dissidents in countries where things like facial identification and metadata scraping are major barriers to posting photos of political protests online.