Compressing the crap out of an mp3?, I want to do almost the opposite of RG Options

[sthayashi]

So back story first:
For long drives, I like to listen to podcasts to pass the time while driving, usually This American Life or Radio Lab. Both podcasts usually have multiple interviews within them, but the interviews seem to be usually mixed at an almost random volume. Plus it seems the volume range between too quiet and too loud is really tight when driving. End result is that I'm adjusting the volume all the time, which is frustrating and probably a little dangerous.

What I'd like to be able to do is just put it on, maybe adjust the volume at the beginning and then not worry about it. Just focusing on the driving and listening.

RG was designed to normalize perceived volume between different song tracks while leaving the actual data intact. I'd like to compress the actual volume within a track, and I don't mind if it's destructive. Is there a good free program that can do this for me? Or is my best hope to decompress->use Audacity-> recompress?

[hlloyge]

Audacity is free program that can do this.
But beware - you could actually do worse than starter if you overprocess the track - leveller effect is a nice start.

[Hanseat]

Levelator is the name of the tool you are looking for.
http://www.conversationsnetwork.org/levelator

It´s the best automatic tool I found for speech in lossless formats. It analyses long range and short range, so different interviews are leveled, but also each interview the different levels of different speakers are leveled. It takes lossless formats only, like WAV, and outputs to the same format.


That´s my workflow:

1. LAME-decoder to WAV
2 (optional denoise and making mono from pseudo stereo)
3. Levelator
4. LAME-compressor for MP3 or Fraunhofer, Quicktime or NERO compress for AAC MP4
5. AACGain for MP3 and MP4
6. Applying tags from original MP3 to final MP3 or MP4

Sorry for germanized english.

This post has been edited by Hanseat: Dec 28 2011, 21:11

[darkbyte]

You may use Stereo Tool for this, there's a command line version. I've made a Foobar setup which uses it in piped setup with my favourite encoder. Although you have to be careful not to use payed options else you will get a Stereo Tool announcment over the audio in the beginning of the track. Of course you can pay for it aswell, it worth every cent imo You can experiment with Winamp or Foobar (using a VST wrapper) in real time.

This post has been edited by darkbyte: Dec 28 2011, 21:47