Need help/advice on capturing audio only from a video clip


I am not sure how to do this, which is why I am asking all you more knowledgeable wooters.

Basically I want to extract an audio only clip from either a divx, quicktime, or a DVD file, and convert the selected audio file to mp3, so I can burn it to an audio CD.

Please tell me the best way to do this.



Use this converter. Cheers!


Get a thing from radio shack, hold on.


sory, if I was on my own comp I wopuld do better, butthis will have to do for now.


Yes Zilla, we know already.


While I’d try Zilla’s cool converter first, an easy and cheap way to do it if you or a friend has an MP3 player with voice record (Sandisk, many flash players), is to play the video clip with speakers on and grab it with the MP3 recorder.


The best way on a computer, other than being able to actually open the file in some professional video editing program and exporting just the audio, is to use your computer to record the sound it is playing. Technically it’s not as good as going through bit-by-bit digitally, but you probably won’t notice the difference, and it will sound much better than holding a microphone up to speakers (especially the low quality consumer grade speakers/microphones you would use).

If you’re on a PC, I recommend using dBpowerAMP Music Converter, unless you need to break one recording into several tracks… if so, skip below to the section on Audacity. Both programs are free. Specifically, what you want to use once you install dBpowerAMP is the dMC Auxiliary Converter. This will let you record whatever sound is playing on your computer. Make sure you close every program that might make a noise except what you want to hear (including things like your e-mail, calendar software, and/or IM), otherwise, if they make a sound, it will be on your CD if you don’t start-over. Before you actually record your sound, click the button that says Test Recording Level. A little box will pop up. Start playing whatever it is you want to record. You should see a bar moving in the box that goes from green on the left to yellow and then red on the light. To the right of the speaker icon is a ramp of bars, like the thing on your cell phone that tells you how strong your signal is. Use this to increase or decrease the recording volume, so that the color bar below is peaking mostly in the yellow; try to avoid going into the red too much (going into the red will distort your sound). Make sure you rely on what you see, more than what you hear, because your speakers can give you a false impression of how loud things are.

Once you’ve got your volume adjusted right, click “End Test”. Click the Options button to make sure everything looks right; the defaults are probably fine. Know where it’s going (probably C:\Converted Music), and make sure that the quality is set at “44.1 KHz (CD Quality)”, “16”, and “Stereo”.

You should be ready to record now. Click on the arrow just to the right of the “Record” button, and select “Record (with Options)”. Make sure you are converting to Wave format. Trying to convert to a compressed format while playing something else that is already compressed may make your computer try to do too much at once, and your sound recording won’t be very good. Make sure again that it is selected on “16 Bit (CD)”, “44100 Hz (CD)”, and “2 ‘Stereo’ (CD)”. If everything looks right click the button at the bottom that says “Convert >>”.

You’ll now see one box pop up on top of another one. The box on top looks a bit like the box you had when you tested your sound. Click “Record” and then start your file. Click it again when you’re finished. You’ll now have a wave file of what you recorded.

If what you just recorded is all you want to be on one track, then everything is fine and dandy. You most likely don’t want to convert your music to MP3, because this will make you lose quality. Because we’ve just recorded in real time what is playing on your computer, instead of going through bit by bit and converting the audio (which may be encrypted if it’s on a DVD or in any other way isn’t ver easy to get to), there could be small errors in the track that you probably won’t hear. Converting a file to MP3 also introduces small errors that you probably won’t hear. The problem is, if you combine these errors, there’s a small chance that you might start to hear these errors. Whatever software you use to burn your CD should do just fine with the wave file. If you’re just going to play it on an MP3 player, then it’s probably fine, but a CD is higher-quality sound than an MP3 usually is.

Things are slightly trickier if you need to break up what you recorded into mutliple tracks. You want to use Audacity. I’m not as familiar with recording in Audacity. All you really need to do is push the big red record button like you did with dBpowerAMP, but getting the sound level right is a little trickier (here you don’t have the green>yellow>red progression to tell you what is correct, it’s all just one color). The line that will stay put slightly longer shows how loud your source has been in the last 3 seconds. If it’s too loud, to the right of that is ‘light’ that will come on that shows if it’s too loud and you’re distorting the audio (“clipping”). It’s just a little trickier to make sure you’re loud enough, but not too loud. Before you record, make sure you check under “File”>“Preferences…”, look at the “Audio I/O” tab, and in the “Recording” box, make sure Channels has “2 (Stereo)” selected. The advantage of using Audacity is, once you have it set right, just click the big red record button, press the yellowish-orange stop button to stop recording, and then you’re all set to trim off the silence at the start and stop and select where you want to make your tracks out of. Before we do that, though, go to “File”>“Save Project As…” and save it someplace on your computer. Now, go to the start of the recording, and find where the sound you want actually starts. Click as close to it as you can get but make sure you’re not clicking any of your sound. Then choose “Edit”>“Select…”>“Start to Cursor”. Click on “Edit” again, and this time choose “Delete”. Now you’ve gotten rid of the quiet that was recorded between when you hit record and when you actually started your sound. Next select where you want your first track to end. Once again choose “Edit”>“Select…”>“Start to Cursor”. This time choose “Edit”>“Split”. This will move your first track down to another timeline. Go through and do it again for every track you’ve got. Make sure you really know what it sounds like where you’re splitting the tracks (I did this once pulling a live show off a DVD, and I thought I saw where the end was, and just went off that, but now when I listen to it it cuts off very abruptly in the middle of the audience applause at the end). Once you have all your tracks separated, select all the sound in each one, and choose “File”>“Export Selection As WAV…”. Read above in the dBpowerAMP section for why you should use WAV instead of MP3 if you’re going to put this on a CD.

Now you’re ready to use those tracks to burn your CD with your program of choice. If this music is supposed to play seamlessly from one track to the next, you want to make sure you select the option to burn along the lines of “Disc-At-Once” instead of “Track-At-Once”. Disc at once will make sure you don’t have any gaps between the tracks that will interrupt the music. It will generally sound better.

If you have any problems or it doesn’t make sense, I can try and walk you through it some more. If you’re on a Mac, I’ve heard you can use GarageBand to do the same thing, and the steps would probably be somewhat similar, but you’d need to ask someone more familiar with Macs and GarageBand for specifics.


If you want to just make a basic recording, use the y adapter, plug it in to the mike in plug and use,
start> programs> accesories> entertainment> sound recorder
to record it.


I think sound recorder only records for 1 minute, unless they changed it on XP. Anyway, I’m assuming what she wants can be played on her computer, and playing it out her speakers and then back into a microphone won’t sound nearly as good. It would be the sound equivalent of taking a picture of your computer screen with a digital camera instead of using the “Print Screen” button. The explanation was really a “sounds good enough only a serious audiophile with the right equipment would tell the difference” level of quality. If she wanted studio-level quality, she’d have to go even higher, but only the DVD would giver her a studio-quality source, so there’s only so good a quality she could get.

(I’m also assuming jurassicjenn means this is a female asking the question)


Seriously, that is the longest post I have ever read. (skimmed)



no, yeah, it took forever to write… but I did learn Audacity a bit better while doing it… i figured i’d do more harm than good giving only a half-hearted answer, so i went all out. i was halfway wondering if I should add pictures…


You should only add pics that have no relevance whatsoever.


heh, that wouldn’t fit this odd, studious-type mood i’m in (check the post i made on the winners gallery).

…studious, that is, except when it comes to studying for my astronomy exam tomorrow, or doing my storyboard, or preparing for my quiz…


holy post…candidate for longest ever 2006.


I thought I was.


Tall already let the cat out of the bag that you exaggerated about your height!


hehe. His MySpace site says 6’1. At least he’s got it right there… :wink:

Actually, at least he’s over 6’ making those claims. I hate it when guys who are 5’10 decide they are 6’1!


I think he claimed to be 6’2", actually…


He said 6’2 here, but he has 6’1 on his page.


Add an inch, subtract an inch. It is all very crucial stuff…when you are claiming to have the longest post as mentioned above.