I now have each band member coming into C3 as audio ports
I also have the main mix out to the PA coming into C3 as another stereo audio port.
This is very good because I have auto record enabled and the recording naming convention set so that it names each file according to the C3 song we’re playing appended with the date and time. This automatically creates a nice ordered file listing.
In rehearsal I want to be able to quickly select a recording I’ve just made and play the “main mix” back. The problem is that dragging the file into the media player just plays the 1st 2 channels. My “main mix” are channels 12 and 13.
The best I have managed is to drag the multi track file into Audacity, and find the main mix channels and solo them.
There may actually be one. If Cantabile is saving audio as multi-channel wave files, there is an easy drag-and-drop way to quickly extract the two relevant channels as a stereo file.
I built another little utility some time ago to help me handle the multi-channel wave files my DL32R mixer creates when recording rehearsals. Since they are a mix of stereo and mono channels, simply extracting all individual channels is not much use to me, so I built a front-end for the tool SOX to de-multiplex multi-channel .WAVs to customized configurations called “profiles”.
Have a look here - and give me a shout when you need more information!
In your case, you would simply need to create a profile that extracts only channels 12 and 13, so it would only consist of one line:
Main Mix, 12 13
Put this line into a text file, call it “Rehearsal Main Mix.txt” or similar and put the file into the Audio Demuxer profiles folder. Now, drop your recorded .wav file into the Audio Demuxer window, select your “Rehearsal Main Mix” profile and click “Process”. You’ll get a shiny new stereo file containing just your main mix. Now you can play it in any media player that plays stereo wave files.
Torsten, you’ve answered both this question, and the next one I was going to ask!
The next question was going to be:- all of the rehearsal files need the same treatment before going into my DAW, which is to consolidate some to stereo, delete others, and rename them. Which your tool seems to cater for. I’ll give it a try. Thanks!
Yup, that’s exactly what I built it for: my mixer records multi-channel wave files with all selected channels included. While recording, it also breaks up longer recording sessions into 2 GB chunks (filename_1, filename_2, …) to avoid compatibility issues. So my tool first extracts the selected channels into mono or stereo wave files and then concatenates them (if needed), so that no stitching is necessary within the DAW. The prerequisite for this is that your multichannel files end in _1, _2, _3, etc.
TBH, the heavy lifting is done by SOX - a great command-line tool to process wave files. All my little tool does is automate the number of necessary SOX steps to de-multiplex, concatenate and name the individual tracks - all file processing is done by SOX.Credit where credit is due
Have a look at the profiles included in the download - one is my rehearsal setup (“Cränk Studio.txt”). You’ll notice that I’ve prefixed the individual track names with consecutive numbers. This way the resulting files are already sorted in the right order in Windows Explorer - I simply select all of them and drag them into my DAW template with the channels already set up. With Audio Demuxer, it’s really just a three-step-process:
drag source file into Audio Demuxer
Select profile and hit “Process”
drag resulting wave files into DAW
This has made creating rough mixes from our rehearsals so easy that I do it far more frequently now - my fellow band members sure appreciate it!
Just tried this - works perfectly thank you Torsten for yet another really useful tool.
Can the output file names be set to contain the input filename? For example
Input File = "Honky Tonk Women_2017-05-18_20-25-51.wav"
Output files = "Honky Tonk Women_2017-05-18_20-25-51-01 Main Mix.wav"
Output files = “Honky Tonk Women_2017-05-18_20-25-51-01 Vocals.wav”
I’ve just uploaded a new version with this feature - you can now select prefix options:
none (duh)
file name (input filename)
date (current date in YYYY-MM-DD format)
custom (enter your own crazy prefix)
The only difference to your request is that the prefix is connected to the file name via an underscore
beware: the custom option doesn’t check for illegal characters (yet), so make sure you don’t enter something that will create funny results like “*test” …
Hi Torsten I tried your new version - thank you for creating it. Questions
With both versions I struggled to get the window to appear - similar problem to that with LivePrompter - I’m using varying numbers of screens, and in this case, using RDP to connect to my music PC, so probably not your normal use case.
The filename prefix worked ok when I restored my profile.txt back to normal. I had put entries in there like
…\process\00-Main Mix,13 14
to get the output files into a different directory. This caused a problem with the path on the new version. I assume I could use the “custom” prefix to do this? What is the format of the custom prefix?
I tried dragging a whole directory of different multitrack song files, (1 file per song) but the tool said it only processed 1 set at a time.
OK, once again, you challenge my simple tools beyond their limits
AudioDemuxer was built to simply create its output in the source directory - keep things simple. So any characters (like leading “…” or a “”) that aren’t “legal” file name characters will have the potential to mess things up. Haven’t built in a filter to remove them, so use them at your own peril!
the prefix will simply be inserted between your original file directory and the file name from the profile, so your “special” file names effectively created paths like “C:\my files\audio\recording\prefix_…\process\00-Main Mix.wav” - which is sure to create something you didn’t intend… So careful with file name acrobatics in your profiles!
same applies to the custom prefix: it is simply a string that will be inserted between the path and the file names from the profile. So you can be pretty creative with it, but do so at your own risk!
Re multiple files: AudioDemuxer will only process multiple files at a time by concatenating them - but that requires them to be
identical in channel structure (no concatenating a 14 channel and a 16 channel recording)
conforming to a naming pattern like file_1, file_2, file_3…
it is sufficient to drag the first file of such a set into Audio Demuxer - it will find the rest.
The tool was built for the recording output of my Mackie DL32R mixer - it puts each multi-track recording in its own directory and splits the output every 2 GB. That’s essentially the usage scenario Audio Demuxer was built for; the big challenge is usually not a large number of different recordings, but rather getting a 2 hour multitrack recording, split into a number of 2 GB blocks, into your DAW in an efficient way.
Haven’t really seen the need for batch processing yet, so yes, Audio Demuxer only processes 1 file (set) at a time. Maybe I’ll get around to that in a next version.
Pretty simple: My DL32R mixer allows me to capture all inputs (up to 32!) right at the mixing desk to an attached USB drive without any need to route everything to my live laptop - which I want to concentrate on providing my keyboards and guitar sounds without being burdened by recording duties.
And I really don’t want to connect 24 channels of audio to my live laptop or studio PC (in parallel to connecting them all to the mixing desk) - currently, my live interface connects 1 input (guitar) and 6 outputs, that’s it.
thanx for sending me your file - I tried to run it through SOX directly and got the error message “WAVE file fmt with padded samples is not supported yet”. I looked at the file header in a hex editor and it does look different from a “classic” PCM wave header - different codec ID, longer format block, includes a “junk” chunk, etc.
I loaded the file with Audacity and exported it again as a multichannel Wave file. That one looked fully standard; and AudioDemuxer demuxed it nicely.
So I’m not sure what file format Cantabile is using to write multitrack recordings; seems like some variant of Wave that SOX can’t deal with - maybe Wave64?
Just to clarify Cantabile’s wave file format… it’s a standard PCM wave file but has an additional “JUNK” chunk to reserve room in case the file exceeds the limits of the standard wave file - in which case it switches to WAVE64 and uses the previously reserved space of the junk chunk for the longer wave64 headers.
It should work with any wave reader that follows the standard RIFF header/chunk format. If not, then that software is taking shortcuts with it’s parsing of the file…
As for addressing @a6325435 the original problem - can you not setup the recorder to record just the two channels you need. If necessary you can create an extrat 2ch output audio port (it doesn’t need to be connected to the sound card), route what you want to record to it and then setup the recorder to just record that one port.
I want to be able to record the main mix as well as all the individual channels. Actually it is more useful for me to record the individual channels and not the main mix, because the main mix only has a very low level of guitar in it - he uses his own amp, which is miced mainly for recording.
Is there any workaround that will let me use 24 bit files? I must admit I haven’t tried the other options - 32 bit and 32bit floating!
Currently there’s no way to record multiple audio output files at the same time - they all get combined into a single audio file with channels stacked after each other.
Is there really a problem with 24-bit files? I thought this was a more general problem with Cantabile’s wave/riff/junk format not being understood by other tools. The easiest fix I can think for this is an option to disable Cantabile’s wave64 support - that way you’d get a more compatible wave file (but it wouldn’t handle long recordings).
Or… perhaps if the other tools understand wave64 correctly, then perhaps an option to always record wave64.
