Q. Could an audio model pick up audio from another application, e.g. amazon music, apply the filters and stream out to the soundcard live? i.e. not reading from an input mp3 or similar. Can this also be achieved where the source is external, e.g. SPDIF input on the soundcard. Is this functionality supported in both linux and windows? I'm using virtual audio cable in windows to create a virtual audio device.
A. Audio Toolbox can open devices that are made available by the OS. It will not create/replace a virtual audio device by itself, but using such a virtual device is a good way to go. To interface with an existent app, you can use either a virtual audio device (set to loop back), a physical loopback, or if the app supports it, an audio plugin format (i.e. VST). Your SPDIF input probably shows up as an input device, so you would be able to select that.
Q. Can such applications be deployed standalone using the appropriate 'coder' toolbox? On both Windows and Linux?
A. You might not even need coder. Look at generating a standalone plugin with generateAudioPlugin. Probably the easiest way to go, as it will take care of the audio I/O for you.
Q. Optional - with the appropriate input mic, I'm using a miniDSP umik-1, can the audio toolbox handle the speaker profiling bit of the process too?
A. I'm not 100% sure what this is, but my first thought would be to look at the Impulse Response Measurer app to see if that does what you need.
If it's relevant, I'm planning on using an ASUS Xonar u7 mk2 usb soundcard in the implementation.
Audio Toolbox will give you the option to use ASIO instead of DirectSound. And if you want to do a test of your device (SPDIF for example), try it with Audacity; if that works, it is expected to work with Audio Toolbox.
