I'd appreciate some help here guys!

Essentially, how do I use Melodyne to ensure that a vocal clip (4 or 5 seconds) is in the same key or scale as the rest of a track? E.G. if I have an exported WAV of a 8-bar loop of the main section of a track how do I determine the key of that loop and, then, use Melodyne to force the vocal to also be in the same key? I guess I could use something like Mixed-in-Key to get the key of the loop...but then how do I adjust the vocal to get it in the same key as well?

Hopefully this post makes sense: understanding music theory is not my strongest attribute.

Melodyne will analyze the vocals for you and tell you what notes are playing.

If you wrote the loop yourself you could just tell us the notes.

Chances are good that whatever the starting fundamental frequency of the loop is, is the key. The tonic(first note of the scale), is often times the first note in a melody.

The other way is to try out each of the twelve tones, and play each of them one at a time while the loop is playing.

Mixed in Key is not 100% so I would not trust it.

___________________

quote:

Originally posted by dj_alfi change your avatar for fucks sake.

First, use Melodyne, your ears (if you have good ear-training skills), and/or Mixed in Key to find the key of the music clip. Then, you'll know the notes that should be in the melody to avoid dissonance. Then, use Melodyne (or your ears, see above) to determine the notes in the vocal melody. From there, you have 2 options: (1) transpose the entire vocal melody up or down to match the key of the music, or (2) transpose individual notes to match the key of the music. The first option is probably preferable to keep the melody the same, but if the key of the vocal is off by a large amount (e.g., a 4th), then changing the pitch may make the vocal clip sound odd/bad.

For example, say you analyze the music and determine that it's in Emin, which uses the notes E-F#-G-A-B-C-D-E, the analyze the vocal and find that the notes in the melody are A-Bb-G-A-F-G-C-D-D. That single flat tells me that the vocal was recorded in either the key of Fmaj or Dmin. Since my instrument clip is in Emin, I know that my vocal clip (assuming Dmin) is one full step (two semitones) lower, so I just need to grab all of the blobs in my Melodyne vocal clip and drag them up two semitones and my new vocal melody is now B-C-A-B-G-A-D-E-E. Done.

Using the second option I mentioned above, you could just change a few notes in the melody to match the key of the clip, which will change the over melody. In this case, you'd raise the Bb to a B-natural, and the F-natural to an F#, ending up with the following notes in the vocal melody: A-B-G-A-F#-G-C-D-D. You can try it, but it's probably not going to give you the results you're after.

If you're working with short simple clips, you should get usable results. But, keep in mind that the results may not always be that great, depending on how long and complex the chord changes in the instrument clip are, and how long and complex the vocal melody is. For example, let's say that your instrument clip is playing a single static chord (e.g., Emin), but your vocal clip was recorded over an 8-bar phrase in Amaj and uses the I-IV-ii7-V7 progression (Amaj-Dmaj-Bmin7-Emaj7), the vocal melody may not sound "right" over the instrument clip, even after you transpose the vocals to the right key, which in this case would likely be down one full step to Gmaj (has the same key signature as Emin).

___________________
cryophonik.com | facebook | soundcloud

Sonar Platinum | Ableton Live 9 | Logic Pro X | Access Virus TI2 Keyboard | Kurzweil PC3X | Nord Lead 4R | NI Maschine

Thank you kindly for the replies.

Melodyne broke the vocal down into individual pieces which I was able to move around (note by note) as needed. I spent a while trying to get the vocal to fit with the groove that I had going. I then gave up, pitched the vocal way down, applied a Grain Delay, made the vocal sound dark as hell, and started re-writing the track as techno haha.

I'll be laying down the arrangement over the next few days

Take a tin bucket. An old one with a little rust. Get some cans. Cut a small hole at the bottom of each can. String some wire though each can. Connect 100 cans (empty PBR cans) on the string, and then tie a knot at each end. Get a string, and hang the tin bucket from the ceiling. Encircle the bucket with your newly fashioned string-o-beercans. Start a fire. Heat up all of the metal. Now you should have an aututuned response from each different can, starting from the first can and going up the frequency range. The bucket is your bass drum, but you can also use it for referencing. Good luck.