0

I'm working on an app which has a requirement for running some basic audio filters (such as normalisation and reverb) on a file. The idea is to take an existing audio file, add the filters, and then write the data to a new file. Crucially, this must be done without any playback and should be fast (i.e. on a 60 second audio file I should be able to add reverb in under a second).

I've looked at several solutions such as The Amazing Audio Engine and AudioBox but these all seem to rely on you playing back any audio in realtime rather than writing it to a file.

Does anybody have examples, or can point me in the right direction, for simply taking a file and applying a basic audio filter without listening to it. I'm sure I must be missing something simple somewhere but my searches have turned up nothing.

Ben Dodson
  • 352
  • 3
  • 12

1 Answers1

1

In general,the steps are:

  1. Set up an AUGraph like this - AudioFilePlayer -> Reverb/Limiter/EQ/etc. -> GenericOutput
  2. Open the input file and schedule it on the AudioFilePlayer.
  3. Create an output file and repeatedly call AudioUnitRender on the GenericOutput unit, writing the rendered buffers to the output file.

I'm not sure about the speed of this, but it should be acceptable.

There is a comprehensive example of offline rendering in this thread that covers the setup and rendering process.

Community
  • 1
  • 1
joelperry
  • 158
  • 2
  • 5