Over the last few days, I registered my copy of MP3 Trimmer. This is an OS X program for making edits to MP3s – splitting, joining, removing portions of them — that does not do a re-encode. This is important, because if you do things like import into Audacity, edit, and then export back to MP3 you are losing quality. There is a reason that when I edit my interviews, I always save them back out as AIFF.
Now that I do the show directly to an MP3 with the Marantz I’ve always avoided like the plague doing an edit to that final product. Besides it being convenient to finish the show and publish it minutes later, I didn’t want to do anything to cause the quality to get crunchy by re-encoding. MP3 Trimmer works by rewriting out entire frames of data without changing them. It might mean that you can’t get the edit exactly on the point where you want it, but that’s a small price to pay.
It’s cheap at $10.95 shareware fee, and is fully operational as a trial, if full of very annoying wait screens. Check it out.
I’m just curious about one thing you said … You said “you can’t get the edit exactly on the point where you want it”. My basic understanding of mp3 is there are about 40 frames per second — each frame is about 25 milliseconds in length. Have you found that granularity to be noticeable? Where one frame edge is just too soon for comfort, and the next one is too late? Again, just curious. I would think ~100 ms granularity would be good enough, but I don’t do a lot of sound editing.
Chris, It’s not a problem I have experienced myself. I just wanted to point out that you don’t have arbitrary granularity. If you had a really precise edit, it might not be possible if the frame boundary doesn’t match exactly what you want. This could conceivably happen if one was trying to edit out a single word or something tight like that. I have done edits that tight in the past, but not recently and not with MP3 Trimmer.
For Windows users, I highly recommend mp3DirectCut. ( http://mpesch3.de/ ). While there, pick up the 1by1 directory player. It’s been my favorite for many years. Both programs are freeware. Hand-coded assembly language against the Windows API produces wicked-fast code with executable sizes that haven’t been seen since computers came with two 5 1/4″ drives. 1by1 can use many Winamp plugins.
The pieces of my first podcast were edited with mp3DirectCut and rendered by playing them in order in 1by1 and encoding the output. Illudium Christmas Vol. 1 was edited and assembled entirely with mp3DirectCut without re-encoding the files. I did Illudium Christmas Vol 2 in Audacity and the rendering time alone convinced me to do my next show by assembling the mp3 files like I did for XMAS vol 1. (Note: I love Audicity and I recommend it when comprehensive editing facilities are required.)
Since it’s neither obvious or particularly well-documented, I’ll briefly mention how to put files together with 1by1. The trick is to open a second instance of 1by1, copy the audio you want (possibly the entire track), and paste it into the instance you’re using to assemble the program. When I finally get my production technique to something I like, I’ll write up a page on it.
There is some selection granularity that may keep you from making totally silent edits. What I’ve done is to move the edit to a slightly different point where it won’t be audible. Another trick is to do a very sharp fade to the edit. If all else fails and you get a click, you can silence it.
Hey Dave,
Thanks for the tip – I used the inverse trim to edit out about 6 seconds of silence caused by dropping a thumb drive on the key board while playing back a pre-recorded interview during the podcast. I kept going becuase I knew I could use this to edit. Otherwise I would have had to restart 15 minutes into the recording…
Yes, I agree. Knowing that you have the possibility for an edit without a full reencode lightens things up and allows you to not worry quite so much. I have to say the $11 for registering this program is not only well worth it but a no-brainer.