Re: [audiolense] Some input appreciated on multi channel room correction
Pascal Roget
Yes, the .wav file is a multichannel filter.
ConvolverVST actually changes its number of inputs based on the filter you load. So, you would load up an instance of the plugin and load your filter. ConvolverVST will then most likely ask to be restarted - when you do, your filter will load automatically and the number of inputs and outputs will match what you selected in Audiolense.
I am not sure how you would map channels to the filter in Reaper. I know I've done this in Cakewalk Sonar and Adobe Audition.
You will notice that these filters introduce significant lag. This can be mitigated by using ConvolverVST's "Partitions" setting.
Finally - you don't load the .wav files into ConvolverVST. Audiolense actually generates config files for ConvolverVST that tell it how to read the filter. You will need to use the drop down menu in the file selection box so you can see all the files in the correction directory, and load up the .cfg file.
We're planning on setting up room correction filters for a 4.1 speaker setup (front L/R, rear L/R and sub). Therefore Audiolense Surround edition has been chosen since that should support the 4.1 config according to the specs.As an audio host we're using Reaper with the the following output channel/track configuration:track 1 - Front Ltrack 2 - Front Rtrack 3 - Rear Ltrack 4 - Rear Rtrack 5 - SubWithin Reaper it's possible to apply a VST filter per track. The idea was to apply the filter which is generated by Audiolense on these tracks with ConvolverVST. However, the filter which Audiolense creates is a single WAV file. That's were my doubts come up:- I guess this WAV filter is a multichannel filter?- How should this single filter be applied to the different channels in Reaper?- Do I need to apply the convolver plugin to every single channel in Reaper and select the same filter for every channel?- If the filter is a multichannel filter, how can I map the channel in the filter to a track in Reaper?Hopefully somebody can point me in the right direction.Best regards,Pieter--
Audiolense User Forum.
http://groups.google.com/group/audiolense?hl=en?hl=en
To post to this group, send email to audio...@googlegroups.com
To unsubscribe, send email to audiolense+...@googlegroups.com
Bernt Ronningsbakk
Hi Pieter,
Pascal covered how this works with ConvolverVST. I Using ConvolverVST or the embedded convolver in JRiver is the easiest way to do this because then you can use the config files to assure proper integration between mains and sub, and use of the right correction filters in each instance.
Base case with Audiolense is that you will get one correction filter for each speaker with your 4.1 setup.
The subwoofer filter is appropriate for use when the subwoofer operates as a standalone speaker. In most instances that would mean when playing the LFE channel. (This needs to be amplified by 10dB outside Audiolense with the current release. In next release this will be handled by Audiolense)
If you want to use the subwoofer to play the lowest notes from each of the main speakers you will get at least one more correction filter in the multichannel file. This filter will have a low pass function that fits like hand in glove to a corresponding high pass filter in each of the main speakers. This “subwoofer offloading filter” is the filter that will give the perfect integration to the mains speakers.
To make things more complicated: There are instances where there will be several subwoofer offloading filters generated for one single subwoofer: Perhaps the user chooses to offload bass to subs at different frequencies, for instance. Big front speakers are offloaded @ perhaps 80 Hz while small surround speakers are offloaded at 160 Hz. That would require minimum two subwoofer offloading filters. Or perhaps the user experiences that the same offloading filter doesn’t work perfect for all speakers, so he just changes the offloading frequency from 80 Hz to 80.1Hz in the front right speaker, since the simulation shows a dip in the crossover region.
Finally, the minimum delay correction, which is possibly better than the “frequency correction”, will generate one subwoofer offloading filter per speaker that uses a subwoofer for the lowest frequencies. So this setup with minimum delay correction will have: 5 speaker filters plus 4 subwoofer offloading filters and you will get a multichannel wav file with 9 channels of correction. If 9 channels of processing is too much for your rig, you can set up all main speakers with the same offloading frequency and simply just use one of the subwoofer offloading filters for all speakers.
You can break the multichannel correction file into mono files by loading the correction file in e.g. Audacity and export channel by channel. Audacity is an open source freeware.
The config files are readable text files. You need to read the config files to see the proper routing of signals. How to read those files are described on the Sourceforge Convolver pages at www. This will also give you a general overview of the mixing & routing that is involved here.
I can’t help you with how Reaper works. You need routing & mixing capabilities for an optimum result. The complicating factor here is of course the sub, which may be asked to play all the channels in the input format, and use anything from 1 to 5 filters in the process. All format channels (“inputs”) that are to be filtered by the same correction filter needs to be added before correction. All corrected signals that have the same destination needs to be added after the correction.
Don’t hesitate to ask if you need further assistance down the road. There may be workaround solutions in the Reaper environment that needs to be used, and there are a few tricks that can be done in Audiolense too, to get the desired result in the end.
Kind regards,
Bernt
_Pieter_
Since I need to apply the correction on realtime audio (no media
files) I think JRiver is no option.
I'm now able to load the .cfg file in ConvolverVST but I can't find
out how to apply the different channels in the filter to the different
tracks in Reaper.
I'll try to separate the filters with Audacity and apply the
ConvolverVST on every single track with the applicable filter and see
how that works.
About the sub: this channel is actually a mixdown of the front
channels. The low/high pass filtering is done on the sub itself so I
guess this reduces complexity a bit.
Pascal Roget
Bernt Ronningsbakk
Just a 4 channel system. Just measure the system with the sub engaged and
you will get the right corrections.
Kind regards,
Bernt
-----Original Message-----
From: audio...@googlegroups.com [mailto:audio...@googlegroups.com] On
Behalf Of _Pieter_
Sent: Saturday, June 23, 2012 1:29 PM
To: Audiolense User Forum
Subject: [audiolense] Re: Some input appreciated on multi channel room
correction
_Pieter_
www.hauptwerk.com) This software takes MIDI for input and plays the
related sound samples. With the help of a VST link I can redirect the
Hauptwerk audio output to a VST host for post processing, like the
room correction.
The suggestion of Bernt to split up the multichannel filter with
Audacity seems a workable route. I've read some info about the config
files and if I'm not correct the config files will look almost similar
after splitting the multichannel filter in separate filters, right?
Front Left:
48000 1 1 0
0
0
C:\Impulses\FrontLeftIR.wav
0
0.0
0.0
Front Right:
48000 1 1 0
0
0
C:\Impulses\FrontRightIR.wav
0
0.0
0.0
Pascal Roget
Audiomulch allows you to link VST plugins together however you want. If you look for screenshots, I think you will understand. It is very similar to Plogue Bidule, however Bidule is more complex to use. If you are serious about digital music production, I think some mileage with either program will be very beneficial. They are similar to FL Studio's new Patcher, if you are familiar with that.
Anyway, in Audiomulch, you set up your sound card outputs (1 through 6) and you drop them into your workspace. You then drop your inputs (the virtual audio cable) which I believe are stereo. Finally, you drag and drop the convolverVST, then load up your filter. The convolver will show up with 2 inputs, and 6 outputs. After that, all you need to do is connect the two audio inputs to the convolverVST's two inputs, and the convolverVST's 6 outputs to the corresponding sound card outputs.
Optionally, you can add a compressor before or after the convolver, and maybe a small VST to control gain/volume if you are using ASIO.
Audiomulch will let you use ASIO or you can also use the Windows mixer.
When I found out about Audiomulch and then Plogue Bidule, this opened up a world of possibilities. In Bidule, you can create a complex scheme, where incoming sound is split up into hundreds of streams that are processed differently. You can then export this layout as a standalone VST plugin!l
Either software, set up as above, will allow you to intercept any stereo or 5.1 audio stream coming from your PC, and apply as much processing to it as you want. You can even install it on a tiny PC that processes the sound coming from the sound cards inputs and spits it back out... like your own hardware DSP.
There are some tutorials on the mp3car.com forums to install Virtual Audio Cable and Audiomulch.
*** PS - actually Audiolense filters will not create "real" 5.1, but I think Audiolense gives you some freedom to choose the volume of the sound that goes to each channel. I am not familiar with the Surround version but keep in mind that you could also use something like the free VST plugin VI Suite and put it in line with your stereo source in Audiomulch, so VI suite converts your stereo signal to 5.1, and then use Audiolense to create a filter to process the 6 channels and apply correction and output 6 channels.
Bernt Ronningsbakk
- 1 out on every channel you can probably use the wav files directly.
Kind regards,
Bernt
-----Original Message-----
Sent: Monday, June 25, 2012 1:55 PM
To: Audiolense User Forum
correction