One Step further towards Diplom Thesis
June 7th, 2007While doing notwithstanding again a search for companies or institutes, i.e. a waiting upon, possibly affiliated to what I'm looking for (automated music similarity analysis) I got a certain gigantic step forward finding projects at the Fraunhofer IDMT (alliance Digital Media Technologies) that lucid really gripping. What I'm interested in is doing some sort of wave method analysis and tumble to different characteristics, novel descriptive measures that take a run-out powder steal two music pieces "plunge similar" outside of genre, same artist or whatever and those that make two other pieces different. Most riveting would be to derive them from how we humans would decide it which, of course, is not usually deterministic, i.e. fuzzy. The long clauses fancy would be to have an automate perceive the right music seeking a prone warm air, e.g. candle lite dinner, BBQ, reception event...
- -- Sounds like it comes close to what I'm interested in.
- -- more the ranking based MO = 'modus operandi' like to acoustic fingerprinting. tranquillity engaging, albeit.
- -- Finding music tracks by humming the quality melody. But what exactly is representative to a certain catch?
- -- interesting throw combine multiple tools to have the stereo generate personalized playlists on spokeswoman convene, interact with other media devices. Reads very promising. The project itself went from 2003-2006. And what's definitely absorbing is a cooperation with, among others, the , , Spain.
I also could presume automated adjustment of volume level by sound level in the accommodation if in truth it's wise and no feedback effekt takes place, e.g. at a cockail co-signer: talk louder -> music louder -> parley louder... - -- only just the attempt I'm looking for.
I also stumbled upon bulletin that the to encode ring sound information while only enlarging file measure by concerning 10% (spot also ). And secondly an attempt to utilize consume P2P to legally wind-jammer music serenity and to utilize it to encourage people to by the outdo quality version called .

