Equalization is sometimes mistaken for a recipe that turns every voice into a polished broadcast sound. In practice, a small adjustment is most useful when it addresses a particular tonal fault in a particular recording. Distance, microphone choice, room reflections and the characteristics of the speaker all influence what reaches the editor. A standard preset cannot take those circumstances into account.
Listen for a particular fault
An equalizer alters selected frequency ranges. Applying a preset without identifying a problem can make a clear recording sound less natural.
Listen at a moderate volume and describe the sound in concrete terms such as boomy, thin or sharp. There is an advantage to doing this before investing more time in the rest of the work. It exposes assumptions that can otherwise survive until the final stage, where they are more expensive or awkward to correct. Allow enough time to examine the result in the same circumstances where it will actually be used. Find a portion of speech where the problem is easy to hear. Keep the first version nearby so that a later change can be judged against it rather than against an imperfect memory.
Consider the original source
Microphone distance, the speaker's own voice and room reflections influence tonal balance. Equalization cannot solve every problem caused by placement.
Approach this stage as a series of small decisions rather than a single correction. Compare the suspect recording with a short test from a different microphone position. When working with more than one variable, change just the one most closely related to the issue you have noticed. A different position, arrangement or passage may introduce an unrelated effect that confuses the comparison. If moving the microphone fixes the fault, prefer that solution for future recording. Make a note of the details that had the clearest influence. This makes it possible to recreate a worthwhile change later without repeating all of the initial experimentation.
Check unwanted low energy
Rumble and handling noise may occupy a region below important voice detail. A gentle high-pass filter can be useful if selected by listening.
Raise the cutoff gradually while playing real speech rather than a silent section alone. Consider what would make someone else interpret this result differently. Is something obvious to you only because you know what happened just before, or because you have spent time with the original material? Look briefly at the problem from the perspective of someone encountering it for the first time. Stop before the voice begins to lose its natural weight. Add clarification if necessary, but avoid inventing a complication where none exists. The most helpful change is the one that addresses a specific difficulty without adding another.
For a related part of the process, see How to Edit a Podcast Without Losing Its Character.
Make specific small cuts
A narrow excessive resonance may produce a hollow or ringing quality. A modest reduction may help more than a broad change to the whole sound.
At this point, a representative example is more helpful than a vague impression about the work. Search for the unpleasant range at low monitoring volume, then reduce it conservatively. Save any useful record of the exercise, including the original version where appropriate. Compare the results in their ordinary context instead of viewing each in isolation. Switch the filter off and on at equal overall loudness. If a small alteration causes a new problem nearby, consider reversing it and trying a more restrained approach. A careful comparison can be completed quickly once the question has been narrowed to something specific.
Handle brightness cautiously
Adding high frequency can make speech seem more present, but also emphasize breath, sibilance and noise. A dull recording is not always improved by a large boost.
It is easy to become preoccupied with one detail while losing the original reason for addressing it. Try a small adjustment and listen to words with S and T sounds. Remember to consider how the choice affects the finished material rather than merely the intermediate result. Keep the change only if consonants become clearer without becoming harsh. Where two approaches appear equally suitable, think about the effort needed to maintain each of them. There may be an advantage to the simpler one if it leaves more room to deal with unpredictable problems. Preserve the comparison so you can revisit it if circumstances change.
Compare different speakers
Each guest may have a different vocal range and recording position. Identical settings can make one speaker balanced and another unnatural.
Evaluate the voices individually before listening to the conversation as a whole. Check whether any detail depends on an arrangement that will change later. It is common to discover that a useful intermediate result becomes inaccurate when the surrounding material is rearranged or another requirement changes. Preserve distinctive voices while avoiding abrupt changes in apparent sound quality. Make a final comparison after those dependencies are stable. In the meantime, leave a clear indication of what still needs to be confirmed. This is more useful than treating provisional work as finished and forgetting which assumptions require another check.
For more on a nearby subject, read Setting Recording Levels Without Clipping.
Avoid chasing headphone colour
A listening system has its own frequency response. Strong corrections made on one pair of headphones may sound strange elsewhere.
Check the edited voice on another reasonable listening system. Consider how this result will appear to someone who did not take part in producing it. They will not have the same knowledge of discarded attempts, temporary adjustments and other circumstances. Substantial improvements should be audible without relying on a single device. A brief independent check can reveal a problem that has become familiar enough for you to overlook. If the check suggests a substantial change, preserve the version you already have and make the correction deliberately. Avoid creating unrelated revisions while trying to resolve one recognizable difficulty.

Keep changes reversible
An equalizer setting is easy to refine if the original file is available. Processing a new version of an already processed file can accumulate unnecessary changes.
Before moving on, establish whether the method can be repeated without reconstructing every decision. Save the original and record the chosen settings in the editing session. Make the remaining checks using the actual output rather than relying entirely on the appearance of the working version. When uncertain, return to the original and rebuild the treatment by ear. If the work needs a further change, keep the reference that shows why. Otherwise, write down the useful choices and put them into practice during the next comparable task. This turns the effort of the first experiment into material that continues to be useful.
Before moving on
Leave an untouched reference. If the adjustment makes the words clearer on several ordinary listening systems without altering the identity of the voice, the correction has a practical reason to remain.
