Mixing Live Tracks - volume issue during transition from track to track

I have been using Audacity to put together Live song medleys. I take a live track from one album and combine with with another live track. The overlap is usually when the crowd is cheering at the end of one song and the cheering that starts before the next song. I just noticed that at times, there is a noticeable difference in volume of that cheering when the new track kicks in. How do I make it a smoother (and quick) transition?

You could try crossfading (see: https://manual.audacityteam.org/man/creating_a_crossfade.html)

OK…i’ll try it. I always thought CrossFading was exactly the same as myself Fading out/in.

Crossfading is when one bit of audio fades out while another bit of audio fades in (the two bits of audio overlap during the transition).

OK, something interesting. I’ve been noticing that volume changes between tracks only when I export the the project into a FLAC file to play in the car or on my phone/headphones. When I play that same transition on the laptop via Audacity (through the project), I do not hear the volume diff…it sounds like a smooth transition. Is something happening upon export???

Audacity Subject Matter Experts…pls help.

Probably not. My guess is that it’s due to differences in the playback systems.
If you’d like to upload the files somewhere (dropbox, google drive, send space, …) I’d by happy to try them out here and let you know how they sound on my system.

Thanks so much…pls send me an email address and and I’ll Google Drive it over.

Just make it a public share, and you can remove the file once I’ve got it.

Got them, and deleted your link for your privacy.
I’ve not had chance to listen yet, so I’ll get back to you later.

I think the problem must be with your monitoring in Audacity. To me, some of the songs sound considerably louder than others.

What I’d suggest is that you downoad and install the “ReplayGain” effect from here: ReplayGain plug-in
Installation instructions are here: https://manual.audacityteam.org/man/installing_effect_generator_and_analyzer_plug_ins_on_windows.html#nyquist_install

  1. Reset all of the track Gain sliders back to zero
  2. Apply the ReplayGain effect to each song individually, using the default settings (you will need to join the split near the start of the first song first). Note that you can use the keyboard shortcut “Ctrl + R” to repeat the effect rather than getting it from the menu every time.
  3. When you’ve applied the effect to each song, mix down the entire project and Normalize to your preferred peak level (I generally go for around -0.5 dB for WAV or FLAC files).
  4. Export the finished mix.
  5. If you intend to keep the Audacity Project, I’d suggest using “Undo” (Ctrl +Z) to get back to separate tracks before saving and closing the project.

Here’s the mix as an OGG file using the method described above: Dropbox - Live Medley 2.ogg - Simplify your life

Firstly, your mix is …PERFECT. It is totally seamless!! This is exactly what I wanted. So If I understood correctly, the ‘replay gain’ is just another parameter that will tell me the volume /gain of a track. You used this to match the gain levels of all the diff tracks which made the transitions seamless.

Can you tell me exactly the right path for where the file needs to be placed within Program FIles for a WIN7 and WIN 10 machine? I can’t see anything called ‘Audacity’.


The “Replay Gain” effect is a kind of amplify / normalize effect, but it also does a clever bit of analysis to work out how “loud” the selected audio is, and then amplifies the selection to bring the level to a “standard” loudness.
By applying the effect to each song separately, each song gets amplified to the same “standard” loudness.

If you are using the current 2.3.2 version of Audacity, use the “Nyquist Plug-in Installer” to copy the plug-in to the correct location.
Then, enable the plug-in with the “Plug-ins Manager”.

I have been using Audacity for about 5 years now (currently using 2.3.2) and never knew there was a ReplayGain plugin for Audacity https://forum.audacityteam.org/download/file.php?id=4685 until I saw this post. I have been using WaveGain all this time. I installed the plugin and tried it out and it works as well as WaveGain or Mp3Gain. As far as I can tell this thing is ready for “Prime Time” and should be added to the plugins that install with Audacity.

The only thing I found missing compared to WaveGain was the DC Offset correction, but low and behold, I kept digging and found The DC Offset Tool DC offset tool (plug-in) by Steve. I married the two in a Macro and now have the same thing in Audacity that I was having to use an outside program for.

Seriously, These two plugins should get more exposure, because they are needed by the people, like me, that transfer (or Download) their music to the digital platform.

Again, Thank You Steve and the entire Audacity Team. :smiley: :smiley: :smiley:

DC offset correction is also available in the “Normalize” effect Normalize - Audacity Manual

In the past there have been discussions about making DC offset correction an effect in it’s own right. Personally I’m in favour as that would make the effect more discoverable, but to date there has been no consensus to do so.

Thanks for the feedback about the plug-ins. I think there could be a few plug-ins on the forum that are ready for general release. I’ll have to try and find the time to work through the plug-ins section of the forum.

So, now that I have it all ready, am I to analyze and normalize each track one by one or do I select all tracks in the project at once?

To use the ReplayGain effect, you must select ONE song, and apply the effect to it, then select the next song and apply the effect (“Ctrl + R” to repeat the effect), then the next song… and so on. Doing this, each song will be normalized to the same loudness.

I described the steps in detail here: Mixing Live Tracks - volume issue during transition from track to track - #11 by steve

many thanks…and what is it normalized to?

somewhere i heard that the target is 89db. Is this true?

what If I have multiple songs on a single row (track)? Will the Normalize work correctly as I would expect there would be many diff volume levels…