So what this is is a sort of audio twitter (but not really). Essentially you can speak into a note for a limit of 30 seconds (though that amount may change in the future). It would record that audio session and it would be pushed to all of the people who were "following" him (you have to approve of someone following you though, not like twitter).
The person can then share the audio clip to facebook or twitter (if possible for twitter).
People on the receiving end of the audio clip get a push notification of a new audio clip. They can listen to all of the new audio clips from everyone they follow all in a row (one playing into the next like a queue) (and are able to reorder the order in which to listen to people's updates).
There needs to be a search function for people to find their friends (including putting a picture as your profile picture to make it easier).
Eventually I'd like people to be able to listen to old posts (before a certain time period in which they're deleted).
Since audio is an expensive medium to keep around (larger data) I'd want this set up so that most of the data is stored locally on their phone instead of in our server (if possible). I'd also like any other architecture tweaks possible to make it less expensive to keep this running.
Additionally, it'll need the ability to input ads before every 3rd audio clip or so (audio clip associated with an image like with spotify).
Again, keeping the cost of running this service is very important.
I've attached general mockups. Disregard the very top one. Only the bottom 6 are what I'm thinking.