i’m trying to calibrate audacity with a focusrite scarlet solo interface and am having trouble. i can’t seem to find a way to get audacity to record any higher level than -6dB. all of the windows and interface settings are at 100% as are the various levels in audacity. the format is 32bit float with is, i believe, the audacity default. seems like i recall reading someplace that that means nothing over -6dB but that just didn’t make sense to me.
what i want to do is to get audacity calibrated such that, when the focusrite reaches clipping at 0dBFS, audacity also comes to 0dBFS. in other words, 0dB incoming signal is recorded in audacity as 0dB, not -6dB.
this is for a friends setup and she is anything but an engineer. i want to create an audacity setup for her that is intuitive in the sense that, if the waveform reaches the top, it’s too loud and that that is something easy to see when she zooms out her workspace. also, if she sees the focusrite blink red, it tells her exactly the same thing and can be checked by looking to see if the waveform reaches the top of her workspace. as things are now, the focusrite blinks red but the audacity workspace shows -6dB, which is not helpful or easy to interpret and trace.
how can i raise the recording level such that 0dB incoming is represented as 0dB in audacity?
the problem here is that the scarlet solo has 2 inputs, one for your mic, one for your guitar or similar. If you only use one of these inputs, the other one will be silent, and when mixing down to mono for a mono recording, you have a simple
as i think on this, it sounds like you’re saying there’s no way in audacity to tell the program to grab each channel as a separate mono file, that the only way is to either let audacity combine them to mono, in which case the signal is divided by 2 for -6dB, or to grab them together as a stereo track. in the latter case, the program audio is recorded at full level and the empty channel can be discarded after recording and then splitting the stereo into two separate channels.
it sounds like there is no way to tell audacity to perform the splitting on the incoming signal. do i have this right?
This is a Legacy problem. Sometimes if you use the manufacturer driver software you might be able to get with your interface, that will let you set up for straight-through mono recording.
Also (speaking as a certified non-window user) sometimes you can dig in the Windows Control Panels - Sound and do the conversion there. This is totally not the first time this has come up.
There are other wacky versions of this. I have a nice interface which will deliver exactly the same volume in either stereo or mono, but neither of them overloads or lines up to “0”.
focusrite scarlet solo
This is the one where the knobs light up red and green for volume measurement, right? Ignore Audacity and pay strict attention to the knob color during the performance. Nobody has to know the recorded volume is 6dB (or more) off and Audacity Audiobook Mastering (for one example) doesn’t care.
Turn the one knob you’re not using all the way down. Select LINE. and I personally wouldn’t touch the “AIR” button if you have one.
Listening to yourself with the wired headphone connection is highly recommended. It’s an easy way to keep track of your announcing volume without constantly glancing at the Solo or the computer. After you get used to it, the announcing volume will be amazingly stable over a long time.
What’s the show? Is this an audiobook? There are some odd tricks.
Audacity will not apply effects, filters, or corrections during recording. However the show arrives, that’s what gets recorded. As above, the performance goes through Windows before Audacity gets it, so sometimes you can change things with Windows settings.
There are some interesting computer problems with recording this way. You can’t easily use the computer for other sound jobs in the off hours. If you decide to Zoom to your sister in Schenectady, the Zoom software may louse up the Windows sound settings, and the Audacity performances before and after the Zoom call may not match.
she’s an audiobook narrator and i’m mastering her materials. the issues come up when there is something to troubleshoot as i’m not where she records. that creates scenarios like this, in which i get files that appear to have minor clipping at the hottest peaks, yet the file she gives me never rises above -6dB.
you are correct about the scarlett solo having the red/green ring lights around the level knob. i’d like her audacity workspace to match the indicator lights, less for her and more for me. i’m spending a lot of time trying to figure out minor problems like this. if audacity was aligned properly to the scarlett, i could move on to the real issues. sounds like i have some digging to do with the scarlett drivers and windows controls now. maybe i can get the signal flow right without making her do more things she might make mistakes with, like splitting channels and discarding things. the less i can put on her, the better.
part of the issue i’m having is that i want her to turn down the level on her scarlett to avoid the clipping. if i do that, though, i want her to turn up the level on audacity since it already has been seeming to be 6dB too low. i haven’t been able to do that, though, hence, the questions. knowing that the incoming levels are being handled together and divided by two clears up a lot. btw, i’m not a windows user either. haven’t been for years.
tell the program to grab each channel as a separate mono file
Audacity can only record from one thing, device, or interface. It can’t do one track from here and the other from over there.
It bothers me a lot that we haven’t touched all the theatrical and studio problems yet. Of the three audiobook standards, home readers never pass noise, closely followed by environment problems.
We’re still at the voice recording stage. What are you going to do if we can’t solve this?
I think the performer should record in full stereo with the track on the bottom (right side) dead flat. And that’s just the way it is. Overloading will happen at 100% (0dB), but probably isn’t going to match the flashing red lights on the Solo. They’re calibrated to warn you ahead of time of potential problems.
They ship the stereo track off to you for splitting, correction, processing, filtering, and etc. The only down side of that will be taking twice as long as shipping mono tracks around.
They are warned not to try to read directly into ACX standard. Read with the Audacity blue waves peaking at (wait for it) -6dB (50%) or a little less.
Whichever way you go, they should read a test with the Solo knobs just delicately flashing red here and there and see what happens. I’m still convinced that you can do a whole show like that and not worry about Audacity and the blue waves at all.
They ship the stereo track off to you for splitting, correction, processing, filtering, and etc. The only down side of that will be taking twice as long as shipping mono tracks around.
You should convert to Mono right away when you receive each chapter (Split to mono with the menu on the left). We assume the performer is keeping WAV copies of each raw recording and a copy of that WAV is what they ship to you. They are carrying the Raw Archive. Keep it somewhere safe. This is the coverage if Audacity goes into the gutter while you’re editing and takes a chapter with it.
“Hello, Elsie? Ummm. Can you send Chapter 14 again?”
Editing in Mono is much easier than in Stereo and although ACX doesn’t demand submission in Mono, they strongly recommend it.
The first and only time you touch MP3 is for submission for publication. You can’t edit an MP3 sound track without causing damage.
What’s the book title? I need to be able to buy it on Amazon now or at least before audiobook submission. There’s no way to work around that.
she’s an audiobook narrator and i’m mastering her materials.
How are you handling reading mistakes? I have seen YouTube recommendations that the performer drop into an edit, insert, and corrections right away when a fluff happens, but I’m not a fan. In your case, the performer is not technically proficient and will need training.
There is another technique. When a mistake is made, the performer pauses briefly, looks back over the script for the nearest even sentence or phrase and reads the whole thing again—we assume without the mistake. Do not stop the recorder. This makes it way easier for you to carefully split the sentences apart and reassemble them without the errors and makes it easier to maintain rhythm, pitch, and emphasis.
This means you will need to do a listen-through to find the errors. There is another trick where the performer claps loudly at the error so make it easier to find.