Sound Finger Print Matching Technology - Ideas Required

Hello,

I am not a programmer but I am looking to hire or ask my team to work on our upcoming project.


The idea require finger print matching of audio like shazam app.


We want the app to recognize what TV channel is currently playing. Can anyone share some ideas on how it is possible? as we wont have live audio finger print of the tv channel... is there any other way?


There are some apps which does that... Mu*fin is one of them... please check video here:





Looking forward to get some ideas.

 

What you're describing, and what Mufin does, is two completely different things. Mufin is a end-to-end system - there needs to be input from the advertisers, the TV/Radio broadcast etc.
What you're suggesting is basically not really possible - granted, you could probably match content to something, but there's no way you would be able to discern which channel is on - programmes aren't distinct to one channel, there's reruns and cross-overs - and there is NO WAY you would be able to do this without a reference - what in the world is the app supposed to compare soundbites to, if there's no reference database?
Not to mention, why are you thinking about this? Ie, what's the idea behind it?

 

Thanks for your reply.

Exactly, there need to be some input and match making of audio to detect channel but what I am confused with is that Mufin have live channel detection and obviously they wouldn't have audio track of live channel programs or event then how does there system detects the live channel? What method they are using?

 

I dunno, but there are plenty of ways to do this - audio-signatures in the broadcast, for one, or just a list of actual live broadcasts at any given time - they could then do a short compare (audio) to check presenters voice, for one, or again, have a audio-signature or something. I dunno how it works exactly, but I'm pretty sure it doesn't work on any show, anywhere, anytime - I'm pretty sure it only works on supported channels / broadcasts.