Modern recording software lets you fix almost anything after the singer has gone home. You can move timing, tune notes, stack layers and reshape tone. That safety net also tempts you to put off decisions that are easier, and often better, made while the artist is still in front of the mic.

Engineer and producer Matt O’Neil, who works out of Lunar Mansion Studios in West Houston, keeps returning to one question: what does this performance need right now? Sometimes the answer is a gear change. Sometimes it’s clearer direction, a better reference, or another take. Often it’s simply to stop recording and protect what already works.

Sort Out Who Is Running the Session

In a vocal session the engineer and producer roles blur easily. O’Neil’s advice is to know which job you’re actually there to do.

If a producer is in the room, your role is to support their vision. A spotless, polished take is the wrong result if the producer wants grit, tension or a looser feel. Before you change the sound or ask for another pass, find out what the producer is listening for and how they like to handle feedback. A quick check-in settles most confusion:

  • Who decides which takes to keep?
  • Who speaks directly to the artist?
  • Which oddities in the sound are intentional?

When the producer is directing, don’t compete with them over the talkback (the mic that lets the control room talk to the singer). Raise technical problems quietly and let the room decide what to do.

With no producer, the artist may look to you for guidance, but don’t guess what they want. Ask whether they need an efficient tracking engineer, honest feedback on their performance, or hands-on help with harmonies and phrasing. If you end up producing the vocal arrangement in depth, say so early. That way nobody is confused later about credits, time or expectations.

Musicians Via production-expert.com

Turn Reference Tracks Into Specific Targets

A reference track only helps if everyone agrees on what they’re hearing. “Make it sound like this” could mean the close, upfront tone, the attitude in the delivery, obvious pitch correction, lush ambience, or how the voice sits against the 808 (the booming bass sound common in modern beats).

Ask the artist to point to one moment and describe what grabs them. Then sort what you can capture at the microphone from what belongs to arrangement or mixing:

  • If a reference sounds huge because of hidden doubles and tight harmonies, a wider reverb on a single lead vocal won’t get you there.
  • If it’s whisper-close and vulnerable, heavy compression won’t add intimacy to a distant performance.

Go after the root cause. If closeness is the goal, work on mic technique and delivery before boosting the high end with a shelf EQ. If a chorus feels small, ask whether another performance would help more than another plugin. Keep the reference handy for sanity checks, but don’t consult it between every take.

Get the Headphone Mix Right Before You Give Notes

Singers react to what’s in their ears. Before asking someone to push harder, soften up or fix their pitch, listen to their cue mix (the headphone blend) and ask:

  • Does the vocal sit comfortably over the beat?
  • Is latency, the small delay between singing and hearing yourself, hurting their groove?
  • Are the monitor effects helping them find the pocket, or smearing the ends of phrases?

O’Neil often uses a subtle plate reverb for space plus a slap delay (a single short echo) for depth. Used with restraint, the pair gives the singer a roomy place to perform without hiding detail.

The mix that makes a singer confident may differ from the one you need to judge the recording. Tuning belongs in this conversation too. If real-time tuning is part of the finished sound, hearing it in the headphones can change how the singer attacks notes. Make sure the artist and producer both understand what’s only being monitored and what’s actually being printed, meaning recorded to disk. Where your routing allows, record an unprocessed safety track alongside the processed one. A flattering monitor mix can also hide problems in the underlying take.

Musicians Photo by Jakub Zerdzicki / Pexels

Give Every Stage of the Vocal Chain One Job

O’Neil’s usual tracking chain starts with a Sony C800G or an AKG C414 XLII microphone into a BAE 1073 preamp. From there it goes through an Empirical Labs Distressor and then a Tube-Tech CL 1B, both compressors. When needed he adds a Pultec EQP-1A, either for smooth top-end air or to tidy up low-end weight.

The gear list matters less than how the stages work together. He tends to hit the Distressor fairly hard on the way in, then uses the CL 1B to smooth the peaks and give the voice a buttery feel. How much of each depends on the singer and how aggressive the performance is.

Whatever you use, each piece needs a defined purpose. If a vocal turns boxy or choked, piling on more compression because the meters are jumping around can make things worse. Before you commit heavy processing to disk, listen closely. Are breaths, consonants and natural dynamics surviving, or is the chain flattening the life out of the phrase? Level-match every A/B comparison so a louder option doesn’t fool you into thinking it’s a better one. If the creative direction is still undecided, protect your options rather than printing a dramatic sound just because it impresses in solo.

The same root-cause thinking applies to cleanup later. Harsh upper mids, muddy low-end buildup and weak presence each call for a different fix, not an automatic routine. Two examples from the source:

  • Mid/side EQ splits a stereo signal into its centre and its sides. It helps with buildup at the sides or resonances in the centre, so the vocal can sit properly.
  • Multiband compression compresses separate frequency ranges independently. It helps control ranges that get excessive on particular notes.

Give One Note at a Time

A trained ear can catch four flaws in one pass: a late entrance, a swallowed syllable, a flat vowel and a shaky release. Reading all four out over the talkback usually leaves the singer overthinking the next take. Choose the one adjustment that will help most, and make it specific.

  • Useful: “Keep your breath supported right through the end of bar 4.”
  • Useless: “Can you do that better?”

Also keep objective problems apart from subjective taste. A digital clip is a technical fault that has to be fixed. Whether a line works better in a breathy falsetto or a belted chest voice is a creative decision. Blur the two and you risk forcing your own taste onto someone else’s record. The aim is to make the next take as strong and comfortable as possible, not to show how many flaws you can spot.

Know When to Comp and When to Re-record

Editing works best when a performance already has clear musical intention. It can’t supply intention that isn’t there. If a strong take ends a little sloppily, an edit can rescue it. If a take sounds tentative or emotionally flat, nudging it onto a grid won’t make a listener believe it.

When you build a comp, a single track assembled from the best parts of several takes, listen across each edit in full context. Individual syllables may land perfectly on pitch, yet mismatched breaths, changed vowel shapes or sudden shifts in vocal weight can make the join sound unnatural. The best technical pieces don’t always add up to the most emotional sentence.

Before asking for another take, say out loud what it’s for. Are you after a stronger emotional read, punchier consonants or tighter rhythm? If you can’t name what’s missing, a full extra take may just add more selection work without adding useful options. Keep the best pass you have, tell the artist it’s safe, and then try the alternative.

Sources

The In-Session Decisions That Actually Make A Better Vocal Recording, www.production-expert.com