Nothing says that converting the MP3 file to a WAV file has to be done in advance; it can certainly be done "on the fly" the first time an MP3 file is encountered.
I am not aware of Windows WAV sounds playing 5 seconds after the event that triggered them (unless the sound chips are busy doing something else). Are we looking for millisecond accuracy, or tenth of a second accuracy? I think the real problem we're having in guessing the correct solution is not knowing more about the application. Are the MP3 files provided at run time, or are they known at compile time? Do the segment time pointers change at run time? How often are parts of an individual MP3 file played? Is the bit rate of the MP3 files known in advance? Is the bitrate constant or variable? Is the program trying to splice together several pieces of a file and get contiguous playback? et cetera ... Rainer _______________________________________________ Delphi mailing list -> Delphi@elists.org http://lists.elists.org/cgi-bin/mailman/listinfo/delphi
