Processing Voice Recordings

GoldWave general discussions and community help
Post Reply
Speaker
Posts: 2
Joined: Thu Feb 02, 2006 12:16 am

Processing Voice Recordings

Post by Speaker »

I have recorded several hours of voice instruction. I am using a USB headset and generally have gotten good recordings. However, because these recordings were done over several days at various times, some parts of the same file are lower in volume than others. I thought the Maximize Volume command applied to the whole file would make all these parts similar, but it seems that it only raises all parts so the one highest peak in the whole file approaches 0 dB.

Is there a command I can apply to the whole file to get the lower volume passages to increase without having to go through the file passage by passage and increase the volume? I have twenty five rather lengthy files to do.
DougDbug
Posts: 2172
Joined: Wed Feb 16, 2005 3:33 pm
Location: Silicon Valley

Post by DougDbug »

There is an auto gain effect included with the optional GWVoice plug-in ($15.00 USD). Of course, you can try it out free.

You could also play-around with EFFECT->COMPRESSION. Compression makes the quiet parts louder and / or the loud parts quieter. TV and radio commercials are compressed to make everything loud.
Speaker
Posts: 2
Joined: Thu Feb 02, 2006 12:16 am

Post by Speaker »

That does it! Thanks for the help. GWVoice was the answer. It took a little experimenting, but I have the right settings now.
Perticelli
Posts: 46
Joined: Wed Nov 09, 2005 2:41 pm

Post by Perticelli »

does the INVERT command switch the low level sounds with the high level sounds? For example, if my voice was very loud and the other person i recorded was very lowlevels and i use invert, would it boost theirs and decrease mine dramatically?
DougDbug
Posts: 2172
Joined: Wed Feb 16, 2005 3:33 pm
Location: Silicon Valley

Post by DougDbug »

Nope! :P

From the User Manual:
Invert
Reflects the selection about the x (time) axis, turning the waveform upside-down. This produces no noticeable effect in mono sounds and has a slight effect in stereo sounds. Inverting a single channel of a stereo sound will produce a simulated surround-sound effect.
It inverts the phase of the signal... It makes the speaker move in, when it used to move out, and vice-versa.

Besided creating a "simulated-surround effect", inverting one channel will kill most of the bass because the left and right woofers are moving in opposite directions. Higher frequencies don't cancel-out as much, due to the shorter wavelengths, random reflections, and some other acoustic-theory stuff.... :roll: Well, some higher frequencies might be canceled at some points in the room...

The Remove Vocals effect also works by inverting one channel. After inverting, the channels are summed (mixed). With one channel out-of-phase, everyting that was identical in both channels is canceled. In fact, the Remove Vocals effect is done by mathmatically subtracting one channel from the other.
...if my voice was very loud and the other person i recorded was very lowlevels and i use invert, would it boost theirs and decrease mine dramatically?
Actually, I don't know how you could "invert" the levels... It's probably mathmatically easy, but the process would probably cause lots of distortion and undesired side-effects. For example, quiet background noise might become like a jet engine!

Compression tends to make the sounds more-equal. The Auto Gain effect is a special kind of compression. Compression can make the quiet sounds louder (louder than they were before compression... not louder than the loud parts.) Or, compression can make the loud sounds quieter (than they were before). Or, both.

Expansion is the opposite. makes loud sounds louder and/or quite sounds quieter. A noisegate is a special kind of expansion. It makes quiet background noise quieter, or even absolutely silent.
Post Reply