Hi,
My understanding is that currently, OpenSL ES doesn't allow decoding of compressed audio (e.g. MP3, Ogg Vorbis, etc) to memory, so if I'd like to do something with the audio in a compressed file, I need to decode it myself with my own software decoder. Now, in theory, OpenSL could be using hardware acceleration on the device for decoding audio, which means that doing the decoding in software would be much slower. In practice, though, does anyone know how common such hardware is, and for which audio formats? Is there any way to find out from the NDK or SDK whether a given device has a hardware decoder for a particular audio format?
(I understand there are licensing issues with decoding MP3 files in particular, which I don't want to get into here... just interested in the technical info.)
thanks
-steve