Linux is everywhere, even on Mars, but if there’s one home appliance I did not expect to find the open-source operating system, it would have been the microwave oven. But Farberware thinks differently and launched the FM11VABK microwave oven running Linux on a Rockchip RK3308 quad-core Cortex-A35 processor to handle Sensory’s TrulyHandsfree voice stack enable voice-assistant features on the Linux microwave oven.
I don’t think I need to go through the full specs of the 1,100 Watts microwave oven and I’ll focus on the voice assistant instead. People who worry about connecting everything to the Internet will be glad to know Farberware FM11VABK does not require an Internet connection or a mobile app for control. Everything is processed locally through TrulyHandsfree technology without sacrificing privacy, and with faster response times.
Sensory offers its “Custom Domain-Specific Voice Assistant” service to any manufacturers of home appliances, vehicle infotainment systems, set-top boxes, home automation, industrial and enterprise applications, as well as developers of mobile apps. So they can train the required words and phrases needed for a given product with up to 150 voice commands supported.
The video below is a voice assistant demo with a microwave oven prototype filmed a few years ago.
Voice commands starting with “Voice genie” wake word allow the user to open the microwave door, set the time, configure the microwave for specific food types, and so on. Sadly, it’s not possible to close the door… I guess it would be possible to implement, but the cost would be much higher with a motor controlling the door. The main advantage of a voice-controlled microwave oven probably just to save you a few seconds of your time each day.
If you’d like to add another Linux device to your collection, Farberware FM11VABK Linux microwave oven is sold for $249.99 on Amazon, but the only user review on Amazon mentions it’s $50 cheaper at Walmart.
Via LinuxGizmos.
Jean-Luc started CNX Software in 2010 as a part-time endeavor, before quitting his job as a software engineering manager, and starting to write daily news, and reviews full time later in 2011.
Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress
This seems to be the sort of application that special microcontrollers are targeted at like the US516P6. A quad core SoC seems like overkill. 😀
This chip looks promising for its price but I could not find any application demo video or anything like this. Do you know any working example video or docs?
Gediz, Look at Rock Pi S
Thank you but I’m actually looking for US516P6 🙂
https://www.bilibili.com/video/av755018430/
I actually managed to get a board with the chip on after months of waiting.. If I can work out how to use it I’ll put some notes somewhere.
It’s quite challenging at this time. Everything is in Chinese, and while several solutions were announced in 2020, most are shown to be “coming soon”. I’ll ask the Wireless Tag guys if they can provide more info.
From what I can tell there is an online app to train new words etc but you need an account with unisound for that.
The chip itself seems to be an Andes NDS32.
I thought I originally saw about it here.. I must have created a false memory from looking at taobao too much.
This one? https://item.taobao.com/item.htm?id=625433843736
I’d say that’s quite the opposite, small increase in accuracy requires substantial increase in compute.
You still need AEC, bearmforming, AGC, etc. Maybe Rockchip offering that code? It not you can extract it here:
https://github.com/freedesktop/pulseaudio-webrtc-audio-processing
AFAIK the Sensory stack does not run on any NPU, it is all implement on the main SOC. This a similar open source package. https://github.com/nyumaya/nyumaya_audio_recognition
When ESP32-S3 ships it will be able to do this.
I’m still waiting for Linux to be available on an electric toothbrush.
what about minix? Minix for mini dildos.
Any voice activated apparatus use is beyond reach to me due to my heavy French accent, although I lived in the US for 30+ years 🙂
No matter which one it always replies “Option not available” or “Did you mean <something>?” where the <something> has nothing to do with what I said, and so on…
It usually ends up with me uttering some colorful language which leads me back to the previous paragraph!
So I guess I’ll still use my hands and fingers with my microwave.
It does not seem to work with a Scottish accent either 😉
https://www.youtube.com/watch?v=TqAu-DDlINs
🙂
A long time ago, we went to Scotland to have software engineering meetings with a couple a Scottish developers: that was a challenge, but we eventually got the information we needed.
I believe human brains still have much more tolerance/adaptation than the current best voice recognition software regarding accents.
Texan deep country accent is also a challenge.
If we can only get Amazon to deliver to the microwave…then you wont have to do crap in life. lmao