Handling drastic changes in volume when recording?

[piblondin]

I'm curious to know how people handle big shifts in volume when they're recording. I recently recorded a play which went from quiet conversations to very loud explosions and gunfire to quiet conversations to loud music etc. I used my Sony ECM-MS907 with AGC on and mic sensitivity set to high. The result turned out decent for my standards, though I would have liked the talking portions to have been louder when I play back the recording. (I'll try uploading something to the gallery.)

More common situations may include lectures which alternate between quiet talking and relatively loud applause or concerts the go from loud music to the musicians' chatter in between songs. It seems impractical to constantly monitor the level on my MiniDisc recorder, especially if I'm trying to be discreet. So, I'm wondering how people capture things like musicians talking to the audience in between songs without having to amplify and distort it afterwards or producing clipping in the loud parts of the recording.

Any advice, as usual, is appreciated. Thanks!

[A440]

Your question got buried by newer ones.

I think you're basically doing all you can do. AGC for speech and theater, a decent level of gain on manual for music. You're right that you don't want to be messing with manual volume during a show. It wouldn't be that effective anyway, unless you have lightning-fast reflexes and perfect anticipation of when people are going to be loud or quiet.

Remember that what you're getting with manual is the way it actually sounded. That's another reason I wouldn't play with the input volume. It's better, as I see it, to have the accurate recording and then, if you want, do some processing on a copy of it.

If your recording is all one track, you could upload it, use Hi-MD Renderer (free from the link in that forum) to convert to mp3 or SonicStage to convert to .wav, and open the converted track in Audacity. Then Select All under Edit and choose Normalize under Effects, which should pump up the quieter parts.

But if you're like me, you make track marks during the event. So you'd have to upload, Combine all the tracks (highight them top-to-bottom, not bottom to top, because it combines in the order you highlight them), and then convert the big track and send it to Audacity.

You could also convert all the separate tracks to mp3 or .wav and then use Nero, if you have it, to burn to a CD. Make it gapless (set gap beween tracks to zero) and use Nero's Normalize function, in a check box somewhere.

I record a lot of concerts, though, and it doesn't really bother me that the between-song chat isn't as loud as the music. As long as they speak into the mic, it's still loud enough.