How to Level Unequal Dialogue in Audacity
Balancing dialogue between two speakers with unequal volume levels in Audacity ensures a comfortable, professional listening experience. This guide explains how to balance audio whether the speakers are recorded on separate tracks or combined onto a single track, using manual adjustments like the Envelope Tool as well as automated processing like Compression and Loudness Normalization.
Method 1: If Speakers Are on Separate Tracks
If each speaker was recorded on their own microphone to separate tracks, leveling them is straightforward:
- Select Track 1: Click the track control panel on the left of the first speaker's track.
- Apply Loudness Normalization: Go to Effect > Volume and Compression > Loudness Normalization. Set the target to Perceived Loudness at -19 LUFS (for mono) or -16 LUFS (for stereo/podcasts), then click Apply.
- Repeat for Track 2: Select the second speaker's track and apply the exact same Loudness Normalization settings.
- Fine-Tune: Play both tracks back together. Use the Gain slider (- / +) on the left panel of either track to make small manual adjustments by ear if one voice still sounds slightly louder.
Method 2: If Both Speakers Are on a Single Track
When two voices share a single audio track, automated effects applied to the whole track can sometimes react unpredictably. You have two reliable solutions:
Option A: Split the Voices onto Two Tracks (Recommended)
- Create a New Track: Go to Tracks > Add New > Mono Track (or Stereo, matching your source).
- Isolate the Quieter Speaker: Use the Selection Tool (F1) to highlight the sections where the quieter speaker talks.
- Move the Audio: Cut the selected audio (Ctrl + X on Windows or Cmd + X on Mac), click onto the new track at the exact same point in time, and paste it (Ctrl + V / Cmd + V).
- Normalize Separately: Once all dialogue from Speaker B is on the new track, apply Loudness Normalization to both tracks independently as described in Method 1.
Option B: Use the Envelope Tool for Manual Leveling
If you do not want to split the tracks, you can adjust clip gain directly on the waveform:
- Select the Envelope Tool (F2) from the toolbar.
- Click on the blue line at the top or bottom edge of the waveform where the volume difference begins. This creates a control point.
- Add another control point right where you want to boost or lower the volume.
- Drag the points upward to increase the volume of the quieter speaker, or drag downward to decrease the louder speaker.
- Switch back to the Selection Tool (F1) once the visible waveform sizes appear roughly equal.
Option C: Automated Compression and Limiting
If manual editing takes too long, you can use dynamics processing to automatically pull loud peaks down and lift softer speech up:
- Select the Entire Track: Double-click the waveform.
- Apply Compressor: Go to Effect > Volume
and Compression > Compressor.
- Set Threshold to roughly halfway between the quiet speaker's peaks and the loud speaker's peaks (typically around -18 dB to -24 dB).
- Set Ratio between 2:1 and 3:1 for natural-sounding dialogue.
- Set Attack Time to 0.20 secs and Release Time to 1.0 secs.
- Check Make-up gain for 0 dB after compressing and click Apply.
- Apply a Limiter: Go to Effect > Volume
and Compression > Limiter.
- Set Type to Soft Limit.
- Set Limit to (dB) to -2.0 dB to prevent any sudden loud spikes from clipping.
- Normalize: Finally, run Effect > Volume and Compression > Loudness Normalization at -16 LUFS to bring the overall finished dialogue to an industry-standard level.