If you watch TV, listen to the radio, or glance at the magazines near the supermarket checkout, chances are there are a few celebrities or politicians that you are getting pretty tired of hearing about. Whether it's Kim Kardashian's wedding or Will Smith's marriage problems, it can all be a bit much. That's why this project from MAKE and Matt Richardson is one of the best things we have seen in a while - The Enough Already!
The Enough Already uses the data emitted from closed captioning, a video experimenter shield, an Arduino, an infrared LED, and a bunch of very clever programming to "listen" to your TV for certain keywords. When it hears something it doesn't like (for example, "Kardashian") it mutes your TV for 30 seconds. If during that 30 seconds, the word is mentioned again, it starts muting again. This is a great project that I can't wait to create for my TV. Awesome work!
Far easier solution: cut off the power cord from your TV.
I'm way ahead of you. We don't have a TV.
What's a TV?
No can do. They don't stream Mythbusters over the web yet.
But they do have MythBusters on NetFlix (depending on your country).
I second!
Agreed, I threw out the TV over five years ago and I feel so much more intelligent and better informed now.
Hah People who kid themselves make up 95% of the population
I was talking with some of the guys from IT after work one day about the possibility of bringing this to the physical world but with voice recognition and bone conductive microphone technology. Just image if you never had to hear that one song ever again...
I hate you.
You're one day too early, my friend.
That is not worthy of code formatting.
...disgrace
What, like a sound recognition device combined with noise canceling headphones?
That's genius!
Quote: When it hears something it doesn't like (for example, "Kardashian") it mutes your TV for 30 seconds. If during that 30 seconds, the word is mentioned again, it starts muting again.
Sorry, but that is not going to work. Not when Deep Space 9 or ST:TNG is on.
You really want to hear a Kardashian taking a hit on the bow from a Klingon and leaking plasma....wait that's almost word for word why she broke up with her last boyfriend.
I lol'd
I love this project
I thought the same thing as OldFar-SeeingArt.
The only issue is that the captions lags behind the audio. So you still hear the names of those people. Perhaps if there was a way to analyze the captions so that you can predict when the names would be said and mute the t.v. preemptively, this would be a better solution. Maybe a Bayesian model would work for this,... hmm.... I'm gonna look this up.
CNN! Hush up Snooki!
http://www.cnn.com/2011/TECH/innovation/09/07/closed.caption.hack.maker/index.html?hpt=hp_bn6
While I love the sentiment behind this project, and it is a fun idea, it would have been a lot cooler 20 years ago. Why?
My "off the air" TV is either going in on RF, or (by way of an HD Homerun) through an HTPC over HDMI. Most of my TV comes in off the Internet by way of HTPC, Apple TV, or Roku, which means the media connection is again HDMI, not NTSC/RS-170.
The change to HDMI, especially in conjunction with the onerous implementation of HDCP, is really stifling innovation. This is what happens when (media industry owned) politicians and lawyers interfere with technology and the marketplace.
I hope someone with the time, energy, and resources, will find a way to make HDMI more accessible to all of us. Whether we like it or not, composite video in home entertainment devices is an EOL technology, and the powers that be would just as soon have the new system be closed to everyone but the big players.
For me, the downside of High Def (as implemented) has largely outweighed the benefits.
My first thought when I saw this project was to modify it to give a TV Tourette's...Take the CC from the video stream, looking for key words or catch phrases, and then play audio samples from an SD card over the TV speakers in place of the normal audio...
For example, instead of:
"Steve Jobs announced his retirement today..."
The potentially unsuspecting viewer hears:
"Steve Jobs [is a !$@# #%$#@$]."
George Orwell was 27 years ahead of his time. But he was wrong about the Ministry of Truth. No need, we can do that in real time.
Everything you know is "Truth". Or at least, now it could be.
Peace, Freedom, and Strength,
Dino
Tele-programmes Department
who's still transmitting the NTSC signal this video shield needs?
Good point...I am not well versed on A/V signaling, but I am guessing that an Arduino would have trouble processing digital video fast enough for a HDMI version of this to be feasible.
If I am wrong, then someone needs to get on it, because I want one!
I believe that the only processing that needs to be done is the text of the closed captioning, so digital is different, but not much more difficult.
For real-time audio processing, I believe a decent ARM processor, such as a TI OMAP (quite portable still), might be able to do it using the closed captioning to significantly reduce the audio-processing needed. Speech recognition is much easier when you have a script.
Now if they can come up with a way to black out the picture for 30 sec when there is a garment malfunction or that beer add you hate.
That is AWESOME. If only there was a way to tell in advance if one of the names were going to be said...
If someone would port OpenTLD/"Predator" https://github.com/zk00006/OpenTLD it would be interesting to see if there is enough power to have it running next to real time on one of the larger (or compatible) Arduino platforms.
Good idea perhaps you could use a speaker output and have the Arduino read the words from the captions and output them through the speaker. Then you could listen to movies/shows in a different language if you change the captions to english for example, this way u don't have to read the captions will trying to watch.
Maybe I can figure out how to wire this to the wife. That way when she calls me by name to do something, it will activate. Then I can then honestly say I did not hear her!!
This reminds me of a project I thought of, but never built, because I never thought it could be so easy! Automatic Commercial Mute. Monitor the text stream, and when you hit mute, the Arduino detects it, looks for the current 3 or 4 words, and maybe 1 or 2 before it, and save a keyphrase that will send a mute when it's detected. You'd still have to mute it the first time for each commercial, but you wouln't have to hear that commercial again, until you ran out of room, and started overwriting the oldest keyphrases. Simplistic, but it could work with some tweaking.
Even better, do it with a TiVo or DVR. Have the Tivo buffer a couple minutes, when you detect the commercial, skip forward the exact length of the commercial.
BTW, you can also do commercial detection by looking for black frames (just before a commercial starts there is a black frame, or at least there used to be), or detecting the volume increase intended to wake up people, but maybe not for long http://usgovinfo.about.com/od/consumerawareness/a/When-And-How-The-Calm-Act-Will-Be-Enforced.htm
My current solution is OldFar-SeeingArt's way, though.
I have one of these shields and it's a lot of fun. It has a little slider switch that controls whether or not to display the television signal along with the TVout signal. The switch could probably be replaced with a dual throw relay so that it could be Arduino-controlled. Then you could temporarily turn off the video signal and replace it with your own message. Just expanding on the April Fool's joke ideas!
WHAT WILL THEY THINK OF NEXT
Could make for a fun April Fools joke. Change the channel any time your keywords from your target's favourite show are seen!
OMG, that is epic. This means that I can overlay data onto an incoming video stream before compression, such as a video camera system? Does this mean we can finally make DIY Security systems?
When have we not been able to make DIY security systems? The necessary components are all pretty cheap.
You probably want to look at LinuxMCE.
Sweet!