Vast majority of media sites play one way streaming. Gaming is “interactive”, but even in that case only small pieces of control commands are sent to the server. And all user interactions are through mouse and keyboard.
Midomi is different that it uses two-way streaming. It is a music site that adds a new dimension of user interaction through the effective use of uplink streaming and voice recognition.
It allows the user to search for music in four different ways:
- Sing or hum a few seconds of the song, midomi will find it
- Say the name of the song or artist
- Type (the traditional method)
- Grab, use a sound source like radio
Sound feeds into the microphone and sent uplink to the server, the voice recognition software dynamically finds the matching song and plays it. According to user feedback, it works amazingly well. I guess you have to try it out for yourself J
Looking beyond its song recognition features (which is cool but not for everybody), midomi can really give us a few hints about the potential power of media streaming not just for content playback, but also user interaction:
Discovery
Just like google not only give “a” result, but hundreds of finds for exploration. This is really provides the foundation for a voice search engine. This new search paradigm creates a lot of new opportunities.
User participation
The model can be adapted to some sort of online contests and shows. Users may also create their own music or sing/mix their sound.
Mobile
Midomi is already supported on iphone. Speech recognition becomes a very powerful tool particularly in mobile setting. We should envision the day when
Check it out, live your “American Idol” dream online.
2 responses so far ↓
Kristina // July 17, 2008 at 1:53 pm |
This sounds neat. I’ll have to check it out. I wonder what other applications could be applied with this concept.
Alex // August 16, 2008 at 12:24 am |
Your blog is interesting!
Keep up the good work!