It's true that based on the bug report linked in the post, the application bundle is modified by a 3rd party installer running as root. I still think Chrome should correctly handle the permissions problem with a warning or asking for root privileges to fix it, but I've updated the post to point out Chrome itself is not the origin of the problem.
Hadn't thought of that! And considering that based on what I calculated in the post he goes off the edge at about 70mph, he probably needs way less that 258 feet right?
You're probably right it's not all real time. I figured it wasn't completely unreasonable to treat it as such because when you watch it, most of it does feel RT and we pretty much see everything that happens in an uninterrupted way.
But for the sake of argument, even if the analysis is skewed a bit, it does suggests they were missing a lot of tracks. So if some of the sequence can be considered RT and some of it not, for them to reach the 88 mph in under 3 miles, probably a lot of it would need to be non-RT, which didn't feel right to me based on the way it's presented in the movie.
Totally agree it would be awesome to be able to figure out what overlaps. Thanks for the comment!
For false positives during a live broadcast, I have the big USB button ready to cancel the light show and goal song. I did not mention in the blog post is that it stops the goal song by playing the Fail Trombone sound (https://www.youtube.com/watch?v=iMpXAknykeg) when you use the USB button to cancel light show
Really not sure how it would work for other sports. The only reason it worked so well is probably because they GOOOAAAAAALLLLLLLL yell is long and distinctive versus everything else. It's not about intensity as much as a pattern in the frequency spectrum. I think it's worth a try with soccer and football.
Mac Mini is hooked up to speakers and TV and it plays the goal song. I set the volume of the media player so that it's just right with the game broadcast so everything sounds good together.
lol well it does go off erroneously sometimes... statistically speaking once every 4 games. BUT... I do have the big USB button ready to cancel the light show and goal song. Something I did not mention in the blog post is that it stops the goal song by playing the Fail Trombone sound (https://www.youtube.com/watch?v=iMpXAknykeg) when you use the USB button to cancel light show. Yes really disheartening, but the one time it happened it made the whole thing pretty funny actually :)
Not a bad idea. However it would probably be hard to make the difference between goals for and against your team without analyzing a lot of what's said prior to a goal.
The model is really specific to the station and commentator. I've trained it on games from a local Montreal station. I tried the other local station with the same model and the it detected no goals at all. So to make it work on any other station, we'd need to rebuild a dataset from scratch.
It's really about the comentator's yell, not the goal horn. So as far as he's excited for Habs goals and not too excited for the other team's goals, it works fine.