I want to do something like speaker verification on iPhone (as a course project). And I'm wondering how to get linear PCM from the speaker. I read about the documentation about queue services, and it seems that it records the sound and then store it to a file. Is there a way to get the linear PCM directly from this? The documentation mentioned something about the buffer, but I don't quite understand that. Maybe it's the key to doing this?
Audio Queue Services is for doing this. Just type in the code you need in the call back function. http://developer.apple.com/library/ios/#documentation/MusicAudio/Conceptual/AudioQueueProgrammingGuide/Introduction/Introduction.html#//apple_ref/doc/uid/TP40005343