To clean audio in Premiere Pro, open the Essential Sound panel, tag the clip as Dialogue and turn on Reduce Noise, without overdoing it. To mix several voices, aim for a shared level: about -16 LUFS for a podcast, -14 for social, with a true peak under -1 dBTP. Rushly Audio does this in one click across the whole sequence: it measures each voice, adds a chain of Adobe effects and sets the level, with AI voice cleanup.
- Premiere can already reduce noise: Essential Sound panel, Dialogue, Repair, Reduce Noise. Don't push the slider: the voice turns metallic.
- Mixing means bringing every voice to the same perceived level. Common targets: -16 LUFS for a podcast, -14 for social.
- Rushly Audio: “Mix my voices” measures each voice, adds DeNoise, EQ, compressor, DeEsser and limiter, and sets the level.
- The mix uses 1 credit per 2 minutes; the AI voice cleanup uses 10 credits per minute of voice.
What Premiere already does for noise
Open the Essential Sound panel (Window, Essential Sound), select your clip and click Dialogue. In the Repair section, check Reduce Noise and raise the slider gently, listening on headphones. Reduce Rumble helps with an air conditioner or a road hum.
For a damaged voice, Premiere also offers Enhance Speech in that same panel. And for heavier work, you can send the clip to Adobe Audition. The best advice is still to record well: no tool gives a voice back what the mic never captured. Badly damaged take? See how to rescue a bad audio take.
Mixing voices: the same level for everyone
In a podcast or an interview, each person has their own mic, distance and voice. The result: one dominates, the other gets lost. Mixing means bringing the voices to the same perceived level, not the same peak.
That's measured in LUFS. Common targets: about -16 LUFS for a stereo podcast, -14 for a social video. By default, Auto-Match loudness in the Essential Sound panel aims for -23 LUFS, a TV broadcast level: too low for TikTok or a podcast.
Also keep some headroom on the peaks: a true peak under -1 dBTP avoids clipping when the platform re-encodes your audio.
What Rushly Audio does
One button, “Mix my voices”, across the whole sequence. Rushly measures each voice, then adds a chain of native Adobe effects to the track: DeNoise, parametric EQ, compressor, DeEsser and limiter. It then sets the level in a closed loop, exporting, measuring and correcting until it hits the target.
You pick the target: Social (-14 LUFS: TikTok, Reels, YouTube), Podcast (-16 LUFS: Spotify, Apple Podcasts) or Loud (-12 LUFS). The true peak stays under -1 dBTP.
With several voices, each track is measured separately and the levels are brought closer together. Music tracks are detected and left alone. An “Original sound | With Rushly” switch lets you compare, without touching any effects you added yourself.
AI voice cleanup isolates the voice and puts it on a new track. The original stays untouched. It sends the voice audio to a voice isolation engine, through Rushly's servers; the video stays on your machine. If it fails, you're refunded and the voice is mixed as is.
The mix uses 1 credit per 2 minutes of voice (minimum 1) and the AI cleanup 10 credits per minute. The cost is shown before you start. Audio is included in every plan, Free plan too.
The limits, stated plainly
Rushly Audio doesn't mix music and doesn't do ducking (lowering the music under the voice). It also doesn't normalize the final mix: it brings each voice to the target. When a track already has settings, Rushly asks whether you want it to take over or you'd rather keep yours.
On a two-person podcast, the matching of the two voices stays light: it was designed first for one voice filmed by several cameras. And if your source is very bad (a clipped voice, a recording from the other end of the room), no mix will save it.
When Premiere alone is enough
One voice, already clean, a single clip: Essential Sound does the job in two minutes. Rushly Audio becomes useful when you have several tracks, several voices, or lots of clips to bring to the same level.
Mix your voices with Rushly
The path in Premiere Pro, once the plugin is installed.
- 1Open your sequence
With your voice tracks. Audio acts on the whole sequence.
- 2Open the Audio tab
In the Rushly panel, Audio tab. Rushly lists the voices it found and ignores the music.
- 3Pick the target
Social, Podcast or Loud. Tick the AI voice cleanup only if the take is noisy.
- 4Click “Mix my voices”
Rushly measures, mixes and checks the level. The credit cost is shown before you start.
- 5Compare
Switch between “Original sound” and “With Rushly” and listen on headphones.
What people ask us.
How do I remove background noise in Premiere Pro?
What level should I aim for with a podcast or a Reel?
Is Rushly Audio free?
Are my audio files modified?
Where does my audio go for the AI cleanup?
Does it work for a podcast with several voices?
The AI editing tools comparison
The 13 tools reviewed, including the cases where another tool does the job better than Rushly. We're not neutral, so we say it upfront.