Digital+Audio

Audiopad I'm not sure how to express my unbelievably, ecstatic joy when I read aboutAudiopad. It is such a fundamental shift in the paradigm of DJ'ing that it boggles my mind.

Once you watch one of thoes videos on the page, you'll understand. (both videos are identical)

Computer assisted mixing has always been too quirky to be useful, hence you probably won't see too many DJ's relying on computers or software. Beat counters are about the most intelligent piece of machinery DJ's carry around.

Enter the Audiopad and its bold features:

  1. Interface: The DJ places "checkers" on the table, in the right places, to use Audiopad. Checkers can represent a microphone, a track, a filter. Moving the checkers around can signify instructions, speed up, slow down, etc. A projector, pointed at the table, shows the DJ the status of his mix. It has an amazingly incredible, dynamic menu system. Sure, we can't all have a projector, but i bet this interface could be converted to virtual checkers moved with a mouse.
  2. Music Library: each track is entirely digital, beatmapped, and can be related to one another. Hence this library organizes itself into like beats, rhythms and can propose appropriate transitions.
  3. Digital Filters: With the interface as dynamic as it is, the DJ doesn't need a million buttons to introduce funky effects on the music he's spinning.

In any case, I'm not a DJ, but I think I know the basics of it and at some point, I'd like to practice. TheAudiopad just knocked my socks off and I had to say so.

Extracting Vocals For a couple of years I've been stuck on the problem of learning how to make my own remixes. I've learned how to patch together other people's mixes, but I've never been able to make my own mixes. The holdup was isolating the vocals of the songs I wanted to use. For example, you'd want to take the vocals of "Dola Re Dola" and put them on to the beat of Nelly, Eminem, or any other stylized dance beat.

My research has revealed that there are two main methods. The first method involves inverting the waveform of an instrumental of the song and mixing it with th regular track. The second involves using a pan/expander plugin. I have yet to thoroughly test the first but it is conceptually sound and more promising. The second one I have tried with moderate success.

Details on the first method can be foundhere. Please note, that the wholesite is good stuff.

More details for the second method can be foundhere, but the basic steps are at the bottom.

I hope you enjoy the following history lesson in the second method.... Winamp, and other audio utilities, have a feature allowing you to remove the vocals, i.e. create and instrumental. How do they do this ? Well, the algorithm assumes that when the song was recorded in the studio, the singer was recorded on the center channel and the other instuments were slightly to the left or to the right. What does this mean? it means that when you compare the waveform of the left and right channels, anywhere it's exactly the same, it must be the singer singing. So to make an instrumental, you just remove everything that's the same in both the left and right channels.

To extract the vocals, you do just the opposite. How well does it work? It depends completely on the song, and the particular recording of the song. Up until recently, I didn't know of a tool to perform this reverse process, until I found thisarticle.

In it, the author explains how you can use Sonice Foundry's "Sound Forge" to extract the vocals. There's a "Process" known as "Pan/Expander". If you set it to "Normal Mix" mode, and use it to apply this process to the entire song, it separates the vocals and the instuments into two different channels. How well does it work? again, it depends on the song and the recording you have. I've just begun to play around with it and I'm assuming using an equalizer to filter out as many instruments as you can before you "Pan/Expand" the track, will help. If it works well, you can expect a few tracks in theAudio section of this site :-)

Summary:

Here's how to do it the easy way in Sound Forge 4.5 and 5.0:

  1. Open the original audio file.
  2. Choose Edit-->Select All to select all the data in the file.
  3. Choose Process-->Pan/Expand.
  4. In the Pan/Expand dialog box, choose the Normal Mix Of MS Recording preset.
  5. Click OK.
  6. Choose Edit-->Selection>Set.
  7. Choose Left in the Channel parameter drop-down list.
  8. Click OK.
  9. Choose Edit-->Copy.
  10. Choose Edit-->Paste Special>Paste To New.
  11. Save your new audio file.

--- A New method that was found here:>>http://gybo.proboards1.com/index.cgi?board=construction&action=display&num=1074422493

Tim G's Pella making tutorial

  1. Get ur mp3 (make sure its atleast 192kps , preferably 256 or more) or WAV (much better obviously)

    1. Load it into soundforge and do a 'PAN/EXPAND' process with the settings set to 'mid side' .. This gives you a WAV with the Left-right material and centre mono material separated..
    2. Grab the Centre mono material (which will hopefully have significantly more vocal than instruments/music.. save it as a mono WAV and load into Cool Edit
    3. In Cool Edit select a section of instrumental with no vox (bit you want to reduce/remove from the vocal) and get a noise reduction profile (at 24000 FFT size + about 300 snapshots in the profile)...
    4. then select the whole WAV.. preview the noise reduction.. listen to what it does..
    5. In the eq graph line bit.. increase the noise red below 200 hz and less between 300-10000hz .. giving a bit more clarity to the main vocal 'region'
    6. Then maybe try some midband compression (300-8000hz) with a very fast attack (with readahead set to about 12ms) and 100ms release.. (i'll have to post a pic of the comp curve to show you how to get good results) to help gate some of the background music left over.
    7. Then go to www.soundhack.com and download the amazing, yet seemingly unheard of,.. spectral plugins (VST) (theres a free trial)
    8. go into cubase (or wot ever prog u use) import the 'pella' you've made and use the spectral gate to reduce non fundamental harmonics (see soundhack pdf for more info)...

The trick is to do this 'in the mix' for best results… Using a fast attack and slow release with the spectral gate helps remove transients like snares and cymbals..

bloody hell .. can't believe i've told you all how to do this.. took me bloody ages to perfect!..

Anyone practicing similar techiques please let me know how u r getting on

I'll post some pellas I've made later to show you the results I'm getting try 6000FFT size with 3000 snapshots on cool edit instead of what i said.. think it works better

Posted: November 19, 2007 | | comments