Midomi. Try to remember that name, chances are it will come in handy when you can't figure out which song is stuck in your head. The technology allows searching for a song by typing the song title or artist as well as by singing, humming or whistling a bit of the tune. The beta site launches Friday and offers search results that include commercially recorded tracks or versions of the song recorded by others who have used the site. The technology also lets people listen to the exact section of each of the results that matched their voice sample.
Melodis, the company behind the site, has licensed 2 million digital tracks for purchasing and currently has a user base of about 12,000. Users can create profiles and rate one other's performances on the ad-supported site. The underlying speech- and sound-recognition technology, dubbed Multimodal Adaptive Recognition System (MARS), differs from similar technologies in that it analyzes various factors including pitch, tempo variation, speech content and location of pauses, said Chief Executive Keyvan Mohajer, who has a Ph.D. in sound- and speech-recognition from Stanford University.
I expect this system to go big because it can really fill a niche. For it to succeed, the technology has to be exceptional, the site name has to be easy to remember (at least it is short) and there can't be any loading issues. Who wants to bet Google buys them out?