DSP Robotics Support

tulamide wrote:Shouldn't the range still kept being normalized (-1 to +1) ?

there can be situations where you want this (FFT, dynamics, displaying things) but it would be a general misconception to look for a proper range when working with float audio signals.

you only have to do it at the end of the signal chain before you go into the IO device driver in your OS.

Rocko wrote:Thus, MS matrixes tend to blend the sides with mid at 'hard panned' input and low setting of 'Sgain'.

that is part of a bigger problem with "widening" in general.

since you dont know how the phase envelope of the input will look like, you can not put up the perfect factor for the mid signal in its releation to the factor of the side signal.

i usually create some fantasy curve for automatic lowering the gain of M when the S gain is raised by the user.

an autogain function is not an option, unless you are in a plug-in and dont mind to require PDC. and it si always a bit inaccurate, too.

KG_is_back wrote:Yes, that is correct. The hard part is the actual detection of common elements in the audio channels. I reckon that from FFT you simply pick frequencies that have common phase element and also similar magnitude. I would have to look much deeper into it and code it properly, which I don't really look forward to at the moment.

general approach regarding frequency: we dont want too much phase difference for bass, and we dont really hear it for the high frequencies. for thje range of about 150-4000 Hz "stereo widening" makes the most sense.

but what is far more interesting is the transients. try out to add "stereo" only for the louder attacks of the input and you will be amazed of the result.