Mixing for Producers: Vocal Masterclass Flashcards
Name and briefly describe 5 steps involved in editing the initial vocal stems before importing them to your song project
Tuning - using pitch correction software to get the desired level of tuning, examining, adjusting and bouncing each phrase to a new track
Editing - removing and reducing visually obvious (unwanted) peaks, breaths and plosives
Timing - check by ear with a metronome for any obvious timing issues and nudge them into a better position. Also check for obvious phase problems
Levelling - Use Waves’ Vocal Rider or manual gain adjustments to get the vocals into a consistent range (don’t remove stereo information by levelling the L-R channels of stereo files to match each other)
Mixer Setup - Bounce the edited vocal track to audio and rename it to show that it has been processed
Name 5 ways to prevent over-processing in the vocal editing phase
Reference against an audio mixdown of your track in it’s current state (and/or with a midi piano playing the chords), not just in solo
Avoid higher values in Pitch Correction plugins e.g. Melodyne’s Correct Key Centre and Correct Pitch Drift controls
Avoid completely removing breaths and vocal artefacts and opt for gain reduction instead
Avoid making edits between words and syllables (in the middle of phrases)
Avoid locking the timing of every word and phrase to the grid or altering timings mid-phrase
In what situation would you NOT want to level vocals during the editing phase
You don’t want to level the L and R channels of a stereo vocal track to match each other
When/why might you actually WANT early and late timings on vocal stems?
Stereo vocal tracks in general, and especially dubs/doubles, can actually increase stereo width imaging by the channels being offset timing-wise from the grid/lead vocal and one another. (This due to the stereo-delay effect)
How would you go about levelling vocals during the editing phase without a specific levelling plugin and is there a benefit to this approach ?
You can level vocals with manual gain automation, a compressor and/or a limiter. This will take much longer but might give more natural sounding result than using a plugin (e.g. Vocal Rider)