How to Fix Common Mistakes in the Studio (part 2 of 2)

by | Jul 28, 2014 | Music Production

“Don’t worry, we’ll fix it in the mix.” We know it’s a cliché but often it’s true. The power of recording in a controlled environment with the right tools and the right engineer is that many common flubs or mistakes can be fixed. So let’s take a look at what can be corrected and what can’t the next time you’re in the studio (see last month’s issue or click here for part one in this series).


For vocalists, we often record 2-4 takes of the song all the way through and then chop up the performance, picking and choosing the best pieces for each section; this is called “take comping.” If we still feel that some parts are weak after we have pieced together a track, we’ll either have the singer go back in for some focused re-takes or we will use some pitch and timing correction.

Now at this point I know what you are saying, “I don’t want to use PCS – that’s cheating.” And this is where it gets tricky for us, too. Most singers don’t want to be told they have pitch problems; it’s kind of the ultimate insult, right? But let’s put it in perspective – no one else in the band has to deal with pitch the same way vocalists do. For most other band members, as long as they put their fingers on the right fret or key, the note is going to come out pitch perfect (fretless string players not included). Vocalists don’t have frets (unless you have T-Pain installed directly in your voice box) so every note they generate can only hit the mark when the singer hears and adjusts their vocal chords in real time – now that’s heavy duty processing!

What we find is that many singers actually learn a lot when watching us pitch correct their vocal performance and we like to think of it as a learning tool that every vocalist should embrace. Many singers actually can’t distinguish flat from sharp, so as they hear that they aren’t quite hitting a note, they actually keep going in the wrong direction getting further from the mark. Because notes don’t last forever, they move on to the next note in the melody, leaving behind an extra sharp or extra flat note. When singers begin to “see” what they are doing as they listen back, they can understand what a flat note sounds like as opposed to a sharp one and with practice they can improve their on-pitch percentage.


T-Pain- and Cher-style pitch correction has actually done a disservice to what PCS is really good at – correcting slight pitch problems seamlessly. What most people don’t realize is that the T-Pain effect is using pitch correction software like a synthesizer – turning the “smooth” knob to 100 and flattening out the voice’s natural vibrato. When PCS first came out, the tools weren’t quite there yet for manual manipulation of the pitch, so a “smooth” knob was included to help mash a highly variable performance back into submission. But now we can zoom into a particular note and actually re-draw it to our liking, keeping intact much of the natural performance and only fixing the parts that went awry. Is this cheating? I guess. Just like it’s cheating for a guitarist to manipulate a performance using a wah pedal or a keyboardist who uses an arpeggiator. We see it as a tool. If you’re recording in a studio, you are already cheating by allowing performances to take place at different times and using plug-ins like reverb to make it sound like you are in different spaces, so why the taboo over using a bit of pitch correction?


When recording keyboards in the studio, we will usually track the MIDI data along with the raw audio recording. You may be asking yourself, “What is the point in doing that if you already have the recorded audio?” There are several advantages to capturing the MIDI, one of which is quantization. Basically…quantizing fixes your screw ups (mostly). Quantizing your MIDI will pull the notes played on the keyboard based on the set BPM to create a perfectly on-beat performance. This can be a great time saver as apposed to cutting/nudging an audio track until the timing is just right. Unfortunately quantizing can also accidentally misread certain grooves, causing notes to get moved inappropriately and can cause performances to sound robotic, so be wary not to over use this tool.

Another great function of having the MIDI data captured is pitch correction. In a live audio take, it can be difficult or even impossible to correct a bad note on the keyboard. Unless the note is isolated, PCS may not be able to correct it. You could overdub the section where the flub occurred, or you could use your MIDI data and simply drag the offending note to the correct pitch. As you can probably guess, this is also a great time saver. And finally, arguably the most powerful use of MIDI is the ability to use a vast library of sample sounds. Not happy with the sound of the studio piano you just recorded with?  Simply replace it with the sample of a grand Steinway recorded in a concert hall!


So what do you do when you’re driving home from the recording session listening to the rough mix and you hear some mistakes or “less than perfect” parts that you’d like to fix? This is one of the reasons why we preach recording as few songs during a session as possible – preferably only one. By focusing on one or two songs during a session, you can ensure that everyone is happy with their parts before the session is over and all the gear is broken down. If someone needs to go back in and record an overdub, the mics are all in the same position and hopefully the overdub will match up with the main take. If you try to overdub a track days later to fix something, you may have new strings, or a different mic placement that can cause the new tracks to sound different. It can be especially hard to re-capture the same sonic qualities of a vocal performance at a later date; everything from attitude, time of day and the number of previous takes can all have an effect that causes that punched-in phrase to sound mis-matched.  Communicate with your engineer about what you feel isn’t quite right about the performance and be as specific as possible.  It might turn out that a bit of digital editing or the use of some PCS is all that’s needed to make that performance shine.


Zac Cataldo is a musician and owner/producer at Night Train Studios, a recording studio in Westford, MA. He is also co-owner of Black Cloud Productions, a music publishing company. Reach him at zac at

Brent Godin is a bassist/guitarist and engineer/producer at Night Train Studios. He is also a talent scout at Black Cloud Productions. Reach him at brent at