Home Vocal Recording and Mixing

How to fix a vocal that was recorded too quietly in REAPER

By StudioRescue Editorial

Published August 31, 2026 · Updated August 31, 2026

Recording too quietly is one of the most common beginner mistakes, usually because the input gain on the interface or USB mic was set low, or the singer sat far from the mic, or the level was kept small on purpose to avoid clipping. The result is a waveform that looks like a thin line and a vocal you can barely hear next to everything else. It feels like the take is wasted.

It usually is not. A quiet recording is not the same as a damaged one. As long as the voice is clearly above the background noise, you can bring the level up to where it should be and keep working. Here is how to do that in REAPER with stock tools, what to expect, and how to tell the difference between a quiet take and a genuinely unusable one.

First, is it actually too quiet?

Before you touch anything, confirm the recording itself is low, not just turned down in the mix.

The one thing you cannot undo

This is the part most guides skip, and it saves a lot of frustration to know it up front. When you raise the level of a recorded take, you raise everything in it by the same amount, the voice and the noise together. You cannot improve the ratio of voice to noise after the fact. That ratio, the signal-to-noise, was set when you recorded.

The fix, step by step in REAPER

Work in this order. Get the level right first, then clean up, then check headroom.

1. Raise the take level

You have three stock ways to do this, and they all get you to the same place.

• Item volume handle. Hover the top edge of the item until the cursor changes, then drag up. This is the fastest way to lift a single clip by ear.
• Normalize. Right-click the item and choose Item processing › Normalize items (peak/RMS/LUFS), or open Item Properties and click Normalize. REAPER can normalize to peak, true peak, RMS, or LUFS. For a first pass, normalizing to a peak of about -6 to -3 dBFS gives you a strong level with a little headroom left. If you would rather match perceived loudness for a spoken or vocal part, normalizing to around -18 to -16 LUFS is a sensible starting target.
• Track gain or ReaComp makeup. If the take is also uneven, a gentle compressor does two jobs at once. In ReaComp, try a ratio around 2:1 to 3:1, set the threshold so you see only a few dB of reduction on the loud words, then raise the make-up gain to bring the whole thing up. This lifts the quiet syllables more than the loud ones, so the vocal reads as louder and more even.

2. Deal with the noise you just raised

Now that the voice is at a proper level, any hiss or rumble came up with it. Handle it gently, in this order, and stop as soon as it is acceptable.

• High-pass the low rumble. Add ReaEQ, set the first band to a high-pass shape around 80 to 100 Hz, and you remove sub-bass hum and desk rumble without thinning the voice.
• Gate the gaps. If the noise is only distracting between phrases, ReaGate can mute the silences. Set the threshold just above the noise floor and use a slow release so word tails are not chopped.
• Reduce steady hiss with ReaFIR. For constant broadband hiss, ReaFIR in Subtract mode can learn the noise from a silent section and pull it down. Use as little as possible, because heavy reduction adds a watery, artificial quality. There is a full walkthrough in how to reduce background noise in a podcast recording in REAPER.

3. Catch stray peaks

After boosting, an odd loud consonant or pop might now be near or over 0 dBFS. Put a limiter last in the chain, such as ReaLimit or ReaComp used as a limiter, with the ceiling around -1 dBFS. This is a safety net for the occasional peak, not a tool for adding more loudness. If the limiter is working hard on most words, back your level moves off instead.

What raising the level will not fix

Adding gain solves the quiet part and nothing else.

• It will not remove noise that was recorded with the voice. You can reduce that noise separately, but you cannot un-mix it from the take.
• It will not fix a take where the noise is as loud as the voice. If soloing shows the hiss and room competing with the words, re-recording with proper gain beats any plugin.
• It will not repair clipping. That is the opposite problem, a take recorded too hot, and it needs a different approach. See how to fix clipped or distorted audio in REAPER.
• It will not remove room reflections or distance. A far-off, roomy vocal sounds louder after a boost but still sounds far off.

When to just re-record

If the solo test in step three showed noise nearly as loud as the voice, or if the peaks are down near -50 dBFS and dragging the take up brings a wall of hiss with it, the honest answer is to record it again. Recording is free, and two minutes of setting the input gain correctly saves an hour of fighting a broken take. It is not a failure, it is the faster fix.

So it does not happen next time

Aim for your loudest words to peak somewhere around -18 to -12 dBFS going in. That is loud enough to sit well above the noise floor and quiet enough to leave headroom for a surprise loud line, so you never clip. Setting the input once, before you sing, is the whole fix at the source. The full method is in how to gain stage vocals in REAPER, and if you are on a USB mic with no interface, the low-output quirks are covered in how to record vocals with a USB mic and no interface.