Is there a plugin that automatically evens volume differences (combining gain, compression, leveling, normalizing as needed, all in one)?
… a (free!) plugin (for Audacity) like the “The Levelator 2” free software program? The Levelator 2 automatically edits all the highs and lows and volume differences in talk files down to a nice, more listenable, file.
The problem with the Levelator is it only works with uncompressed…
Not a big problem but I am importing Mp4 video into Audacity to get the audio, then exporting as wave, running through Levelator then back to Audacity to get a preferable Mp3 file.
Just thought this would be an awesome plugin/effect, if it exists. Levelator does not disclose exactly what it does other than somewhat referring to it being a culmination of compression, leveling, and normalizing, and for talk only… which I know I can do in Audacity but I’m never quite sure what settings are best; and they seem to vary depending on the material. In other words, the Levelator program is smarter and simpler.
“Mp3Gain?”: Only normalizes
“Mp3DirectCut?”: Normalizes and enables more specific envelope type editing but not as good or user friendly as Audacity.
I also searched Audacity’s websites’ links to plugin sites and did not see anything.
Chris Capel designed “Chris’s Compressor” to do exactly that. It simulates the audio processing in a broadcast radio station. You can put widely varying volume segments in and it sets them all to the same volume.
Chris’s expressed purpose was to make it so he could listen to opera in the car.
When I use it on a podcast, I increase the compression from 0.5 to 0.77 and the result is indistinguishable from the same show on the local radio station.
Clue in the title : it’s designed for conversations where there is a big difference in the volume of the different persons speaking. Not sure if it will produce OK results with music , maybe too abrupt , ( Chris’s Compressor is a better-bet if it’s music ).
I’m kind of surprised that there’s no built-in graphical C/L kind of tool.
In my dreams… I could draw a horizontal line on the audio track and call that “max volume”, then draw another line and call that “min volume”, and then have the audio content in any selected region rescaled appropriately so that no signal is ever lower than min, nor louder than max. Dynamics otherwise preserved, just scaled to the range that I want. This is kind of what a C/L does, but I find the available C/L tools touchy & time consuming to tweak to achieve my actual goal. I seem to spend a lot of time hunting for the “sweet spot” and getting either too much amplification of the louder parts, or too little amplification of the quieter parts. Maybe if I spent more time practising and really trying to understand the C/L plugins I could improve my skills, but ya know how it is, just trying to fix the darned audio so I can rush the mp3 to the radio station for airplay TOMORROW, as usual.
I did in desperation try extracting a mixdown wav and shoving it through Levelator. The result was brutal but pretty effective… Unf it destroys deliberate effects like intro music fadeouts and outro music fade-ins, etc. It’s kind of a sledgehammer. I think if I had it all to do over again (my current project) I’d run my raw wav files (straight off the recorder) through Levelator before they ever got into Audacity, then refine them in post.
(Background info:) I’ve been struggling for days with a nightmare interview recording in 4 tracks (1 stereo 2 mono) where I used a Zoom H4N for a 3-way chat, with me on the cheesy builtin condenser mics and my interviewees on decent Shure cardies. The only available room for the interview was too bright… So my guests sound clean and fairly well isolated but my own track is a friggin’ nightmare of natural reverb, background noise events including their voices and feet shuffling, background hiss, all kinds of ugly. I don’t ever want to do that again. The condenser mic on the H4N is just so tonally different (and different in response and every other way) from a decent vocal mic that it’s not worth even trying to use it imho. Anyway, I’ve invested in a PodTrak so everyone can have a proper XLR mic. I have enough Shures. Heck, I could afford to buy a couple more for the hours I’ve put into trying to fix this recording
If by “C/L” you mean “command line”, then that’s a contradiction of terms isn’t it?
“Command line app” generally means a app that you run through a terminal / terminal emulator window “without” a graphical interface.
What happens to silences, or parts of the sound that are very close to silence?
If there is “absolute silence” then it can be amplified as much as you like and it will still be absolute silence. “Very close to silence” is likely to be just noise, so amplifying that up to the same level as the other sounds will give you a very loud “SHHH…”
That’s the problem right there. If you start with a high quality recording, you are already close to your goal so it’s relatively easy and quick to get there. If you start with a nightmare recording you are a long way from your goal and it is a long and difficult journey to the goal - if you can get there at all.
Yes, C/L = “compressor/limiter” – I used to have a few of them in 19 inch rack mount format… ahh, the pre-digital days! For cmd-line abbrev I would use CLI…
Well to tell the truth I do heavy noise reduction before I even get to other sound quality fixes. I pull background hiss right down to zero wherever possible. So I wouldn’t go to amplitude shaping until I’d tamed the baseline noise. But anyway. Some more graphical interface (drawing right on the audio track) w/b nice for various tools including Fade In/Out, C/L. Just my $0.02 of course.