How to transfer big-sized video files from Google Glass, to be viewed few minutes after the capture? - google-glass

I have to transfer large-sized video file from glass to a server or on a local computer, just after capture (~400Mb), to be viewed as quick as possible on the target device.
I tried to use the mirror API to get the video but the upload time to google platform is really high (more than one hour for a 10-minutes video length).
I also tried to make an app to capture video and send it over the wifi, but it took about 3 minutes for 18Mb (So, more than one hour for 400Mb).
We also studied the android Wi-fi P2P solution, but it seems not available on Glass at this time.
And I tried to launch Droid NAS or AirDroid application to access Glass file system over the air. Application installation is OK, but I can't run them.
Has someone any solution/recommendation ?
Thanks in advance,
Julian

Since you're willing to do it to a local computer, you can just plug Glass into the local machine and transfer the file. Glass is treated as a Camera device, and most PCs and Macs have software that will let you transfer the videos directly over.

Related

Unreal Engine Pixel Streaming

I have set up a pixel streaming in UE5 in AWS(Amazon Web Services) it works perfectly on my laptop, mobile and my colleagues system (same WiFi).
When I send that public IP4 link to my client in another city it just appears blank, after clicking the triangle (play button) in Pixel Streaming.
Edit 1: when I change the location(IP) using VPN, it also not working on my laptop too.
EDIT 2:
I'm using UE5 pixelStreaming, It works on many devices/locations, but it didn't work on some locations. I contacted AWS the checked different locations and its working on their side.
surprisingly i didn't do anything it worked on all devices in all locations the next day.
On the next day this problem arise again it didn't work in my client's whole office.
after a few hours, it started to work on some of my client devices(mobile /web browser).
I checked the remote desktop I find out that webRTC data is not coming there.
Why WebRTC data not coming in some of the devices?
I think it might be because you didn't allow access from other networks on AWS, so it only allows those on your network to see it. While not this exact issue, something similar happened to me before.

Extremely slow S3 download from only one device

My download speed from a specific AWS S3 server is extremely slow - but it only happens from my Desktop PC, connected through Ethernet.
If I use my mobile phone or any other device (connected through wifi), the download speed is perfectly fine.
Basically getting less than 300Kb/s instead of 10Mb/s+
Any ideas on this? It's really weird...I've tried using Cloudflare Warp and similar stuff, nothing worked.

Can I run Dragon NaturallySpeaking on Amazon WorkSpaces?

I use Dragon dictation software including by using it over a remote access to my home computer. I also once had it set up on the network server computer at a client site which I could also access when logged in remotely. But now, I have a situation where my access to work info is over Amazon WorkSpaces. Does anyone know if Dragon Pro can be installed on and accessed through an Amazon WorkSpace desktop?
TIA!!
Your question was bugging me for a long time, finally I gave in and rented a WorkSpace and tried Dragon.
Recognition for me is totally fine as long as your local computer's standard audio input device is actually the one you want to use Dragon with. If your default input device is the built-in micro of your laptop, computer or screen, recognition will most likely be bad. I used a Sennheiser MB Pro 2 with a USB dongle on a Windows machine for testing, as well as a TravelMike with a USB MultiAdapter. Both work fine.
As for factors that influence recognition quality:
I know from other virtualization/remote desktop solutions that the codec that transfers your speech to the virtual/remote machine may work well for one microphone, and not for others. Try a different mike and see if that improves your recognition.
In order to rule out microphone issues, I suggest trying different internet connections (WiFi, Ethernet, tethering, different internet providers). Missing words from speech recognition over remote connections can indicate unstable or slow internet.
As far as I know, there is no way to patch a USB device through to the remote WorkSpaces machine. What you could try is install a VPN on the remote machine and use your smartphone as a remote microphone in Dragon. Your smartphone also needs the VPN. If you get that to work, recognition should be as good as on your local network, given that internet connection is fast and stable enough.

Which AWS EC2 instance type is most optimal for audio streaming?

I'm in testing stage of launching an online radio. I'm using AWS CloudFormation stack with Adobe Media Server.
My existing instance type is m1.large and my Flash Media Live Encoder is streaming mp3 at 128kbps which i think is pretty normal but it's producing a stream that isn't smooth & stable at all and seems to have a lot of breaks.
Should i pick an instance type with higher specs?
I'm running my test directly off of LiveHLSManifest link that opens on my iPhone's Safari and plays on browser's build-in player..which doesn't set any buffering on client side - could this be the issue?
Testing HLS/HDS links directly on iPhone's Safari was a bad idea. I relied on built-in players already having some sort of buffering configuration by default but noo... I was able to receive stable & smooth stream when i used players like Strobe Media Playback, FlowPlayer etc.. Hopefully, this answer will save someone some time.

Hosting a web service on a microcontroller

I saw some devices online which had wifi and exposed some information which was collected by an samrt device (iOS in this case). Do you know of any microcontrollers which would be useful to expose some data read from a sensor (by WiFi)?
Example: Expose termometer web service - if someone request for the current temperature make the reading (or pass the cached reading) and send it to the requester.
Thx for your help in advance.
The Raspberry Pi device appears to be available for cheap, has an Ethernet connection, and even has a beta Stack Exchange community. You may be able to do something cheaper -- no GPU, for example -- but this might be a reasonable starting point.
I have recently played with a http://ruggedcircuits.com/html/yellowjacket.html to do what you want to do.
I think Yoctopuce have some hardware that might allow you to do exactly that:
http://www.yoctopuce.com/EN/article/an-autonomous-solar-weather-station
This is a solar weather station, based on a temperature sensor on a WiFi module. The WiFi module reads the temperature from the temperature sensor and automatically post it on the internet. It's fully plug and play, no programming needed.