Vocal Removal Plug-ins
This page is about techniques for removing the vocals from music recordings, or more generally removing or isolating one part of a recording from the rest. Unless the original multi-track recordings are available, this is not generally possible. That is, there is no universal, reliable way of separating what you want from the rest. There are however some tricks you can apply (on stereo tracks only) that can fool the ear into thinking that the vocals have been removed or isolated. Whatever technique works best on your particular track, and how good the result sounds, depends entirely on where the vocals are positioned in the stereo field, and on the vocals' precise frequency and overtone content compared to the rest of the audio.
Case 1: Vocals in the middle, instruments spread round them
If the vocals are panned in the centre of a stereo track (fairly common in "pop" music tracks), the so-called "vocal removal" technique can sometimes be effective. This removes what is common to both tracks (i.e. the vocals), leaving behind what is different (i.e. the instrumentals).
To try this technique in Audacity, you split the stereo track into its left and right channels, make both mono, invert all (or a selected part) of one of them, and play back the result. See this page of the Audacity FAQ for step by step instructions on this technique.
You should note that this will remove everything panned in the centre, not just vocals. In pop music this can mean removing the bass and rhythm parts. Removal of the vocals is normally incomplete and will leave artefacts behind, especially where reverberation (echo) is used, and where there are backing vocals. Sometimes certain minor strands of the instrumentals are exposed which are completely covered in the uninverted mix. This technique work particularly badly with tracks where a lot of stereo reverberation has been applied, as this spreads each sound source, and makes them very hard to extract from each other.
Case 2: Vocals in one channel, everything else in the other
If you have an unusual stereo track where the vocals are mixed hard into one channel and everything else hard into the other channel, you can simply split the stereo track into left and right and delete the vocal channel. If you want to isolate the vocals rather than remove them, you can of course delete the other channel. To split a stereo track in Audacity, click on the downward pointing arrow at the top of the Track Panel (where the mute/solo buttons are ) > Split stereo track. Then to delete one of the channels, click the [X] to left of the downward pointing arrow. If you delete the wrong channel, use Edit > Undo to get it back. Finally, click the downward pointing arrow again and choose "Mono", so that the track will play out of both speakers.
As well as the above methods which are supported in Audacity itself, there are various third-party plugins that can be used to try and remove or isolate vocals.
Center Pan Remover
There is a Nyquist plugin called "Center Pan Remover" which can automate the steps involved in the Case 1 above so you just have to select the area of track you want to apply the effect to and click Effect > Center Pan Remover. This can be downloaded from the Audacity Nyquist Plugins page. This plugin also has an option to merely invert a range of frequencies in one channel rather than invert the whole channel (and so may be less destructive of the non-vocal parts of the music than a simple inversion of one channel). Note that although two channels of output are produced, the result is mono because both channels will be identical.
To install this plugin, place it in the Plug-Ins folder inside the Audacity installation folder. On Windows computers, this is usually under "Program Files" The plugin will be available after restarting Audacity, underneath the divider in the Effect Menu.
Three Windows plugins are known of that can be used in Audacity for vocal removal and isolation.
kn0ck0ut can sometimes remove vocals where they are not centre-panned but are rather different in frequency make-up compared to the non-vocal parts of the track. This is a free VST plugin. It works on two mono tracks (split from a stereo track and made mono as above) by extracting the right channel frequencies from the left hand frequencies, leaving the "result" in the former left channel. Alternatively, if your vocals are centre-panned, this plugin can often make a good job of isolating them (as opposed to removing them).
Voicetrap is a commercial VST plugin and features centre channel removal by frequency-based and cepstral methods (i.e. not a simple subtraction of one channel from the other) and (uniquely) provides stereo output for the "vocals removed" track. It also offers the option of isolating centre-panned content. There is a demo version with a "mild" vocal removal setting (but no isolation ability) that works in Audacity.
If you wanted to try either Kn0ck0ut or VoiceTrap in Audacity you have to to copy the unzipped .dll file from each plugin's zip file, along with the unzipped Audacity vst-bridge.dll from this download http://audacityteam.org/vst/vst-bridge-1.1-win.zip
into Audacity's Plug-Ins folder. On restarting Audacity the new effect will appear in the Effect Menu underneath the divider.
Note that all VST plugins in Audacity only have a generic tabular interface due to licensing restrictions imposed on us by Steinberg.
Should vst-bridge.dll not work, try the previous VST Enabler for Windows: http://audacityteam.org/vst/VST-Enabler-0.1.3.zip
and just place the unzipped enabler in Audacity's Plug-Ins folder along with the plugins' own .dll and close and restart Audacity.
OS X Plugins
On OS X you can try theAudio Units plugin (14 day free evaluation) in Audacity, as long as you use Audacity 1.3.2 or later, for which you must be on OS X 10.3 or higher.
This is rather similar to the Nyquist Center Pan Remover in that it aims to reduce the strength of centre panned vocals but is slightly different in that it has a phase offset adjustment and a slider for adjusting the intensity of the vocal reduction.
Try putting the voxreducer.component in either of these two locations as Audacity should see it there
If that does not work you could try either of these locations (as suggested by the program documentation)