Hello all and thanks for allowing me in.
My question relates to and old recording of a band made on a cassette recorder.
I have used Audacity to digitize it, and have used the amplify effect so far.
However the recording was quite bad. Lot’s of echo, not much detail, hard but not impossible to hear drums, voice audible but also not great. For anyone who knows Festival Hall in Melbourne you will know what I mean.
I know I will never end up with outstanding or even excellent sound, but are there any common tips on things to try to at least improve things a bit?
Cheers and thanks again
More info, Windows 10 and Audacity version 2.2.2
Did you want a sample of the recording? This may help with suggestions
It’s almost impossible to remove reverberation. It sounds like the recording quality is too bad to be able to make any significant improvements.
The current version is Audacity 2.4.2. It is available via the Audacity website: Audacity ® | Download for Windows
I’m not very hopeful, but yes you may post an audio sample. Just a short section in WAV format. See here for how: https://forum.audacityteam.org/t/how-to-post-an-audio-sample/29851/1
You can try experimenting with the [u]Graphic EQ[/u]. The main vocal frequencies are in the lower-mid range and boosting the higher frequencies will bring-out the cymbals and “T” and “S” sounds. But, just experiment… There is a LOT of overlap between ALL of the instruments and vocals so you can’t really “isolate” a particular instrument or vocal but sometimes you can improve the overall sound.
There is a general rule that it’s better to cut with EQ than to boost and you might just try bringing-down one slider at a time (maybe -9 or -12dB) to see if anything helps… That can help if there are frequency bands with excessive ringing/resonance.
If there are any excessively-loud or excessively-quiet parts you can fade-up or fade-down with the [u]Envelope Tool[/u]. And the “trick” is to fade with no sudden changes. You also might want to amplify (or normalize) the songs independently (or you might want to keep the loud songs loud and the quiet songs quiet). If you do that… Assuming there is applause/crowd noise between songs and assuming you are keeping one-long concert-length recording you’ll need to re-join the songs with crossfades, again so there are no sudden-unnatural changes in volume.
After using the EQ effect it’s a good idea to run the Amplify (or Normalize) effect before you export. If you end-up boosting anything you can push the levels into clipping (distortion) and either of those effects will bring the level safely down if necessary.
\
Depending on how much time you want to put-in… With live recordings I’ll edit-out any excessively long gaps between songs as well as any “excessive” talking by singer or band members. Often when I’m editing, I’ll “steal” applause from one part and blend it in somewhere else. I always use crossfades so you can’t tell it was edited.
I’ll do that a LOT when I’m making individual song files because sometimes the next song (or the next song introduction) starts too soon, or there is other talking and I can’t make a natural-sounding applause fade-out at the end of the song without stealing it from somewhere else… I usually like a 1-2 second crowd noise fade-in at the beginning of the song and a nice-long 10-15 second fade-out at the end, depending on what I’m starting with.
For anyone who knows you will know what I mean.
I’ve never been there but the amount of reverb that sounds great coming from all directions in a big room sounds artificial when you hear it from a pair of speakers in your living room. Usually the microphones have to be positioned close to the stage (or close to the individual instruments) for more direct sound and less reflected sound.