Is there an existing way to use a mic to measure a woofer's performance and then develop some DSP filters to correct for under/over-shoot of the cone at various volume levels?
I've been thinking about the servo feedback for subwoofers and it seems like with a mic we can build a model of each woofer's performance across the FR and at different volumes and adjust the signal preemptively to compensate for its issues. No need for a live feedback loop. Could apply to mids as well.
By amplifying the signal in the right way at the right times there would be much less need to carefully choose the right Qms, Qes, Qts, Mms of woofers. Any reasonable speaker could get pretty close to perfect cone control, all without a live feedback loop.
I've been thinking about the servo feedback for subwoofers and it seems like with a mic we can build a model of each woofer's performance across the FR and at different volumes and adjust the signal preemptively to compensate for its issues. No need for a live feedback loop. Could apply to mids as well.
By amplifying the signal in the right way at the right times there would be much less need to carefully choose the right Qms, Qes, Qts, Mms of woofers. Any reasonable speaker could get pretty close to perfect cone control, all without a live feedback loop.
"Driver linearization" sounds like a good term to describe what I'm talking about. I can't tell for sure if AudioVero is doing what I mean, but this is a term I found in their documentation and it seems likely to be what I'm looking for.
"Phase linearization" is a similar term but that's not what I'm talking about.
While doing some googling around this I found some papers that talk about using a Volterra model to build an inverse non-linear filter to create "pre-distortion" to add to the signal to offset the driver's distortion resulting in a more linear output with less deviation from the original input signal. That is exactly what I was wondering about, but has anybody done this in a free tool like REW and rePhase?
Papers:
Realtime Loudspeaker Linearization
Electrodynamic Loudspeaker Linearization using a Low Complexity pth-Order Inverse Nonlinear Filter
On the linearization of a parametric loudspeaker system by using third-order inverse Volterra filters
AudioVero looks pretty neat but I'm not going to pay $400 just for an experiment. Also, I'm using Linux and not Windows and their screenshots only shows Windows. Thanks for pointing them out though!
"Phase linearization" is a similar term but that's not what I'm talking about.
While doing some googling around this I found some papers that talk about using a Volterra model to build an inverse non-linear filter to create "pre-distortion" to add to the signal to offset the driver's distortion resulting in a more linear output with less deviation from the original input signal. That is exactly what I was wondering about, but has anybody done this in a free tool like REW and rePhase?
Papers:
Realtime Loudspeaker Linearization
Electrodynamic Loudspeaker Linearization using a Low Complexity pth-Order Inverse Nonlinear Filter
On the linearization of a parametric loudspeaker system by using third-order inverse Volterra filters
AudioVero looks pretty neat but I'm not going to pay $400 just for an experiment. Also, I'm using Linux and not Windows and their screenshots only shows Windows. Thanks for pointing them out though!
Last edited: