Browse all docs
Remove background noise and echo from a video's audio
PandaStudio denoises a clip's audio on-device, removing background noise and, optionally, room echo, so voices sound clean without sending audio to a cloud.
Enhance voice, in the Audio tab, cleans a clip's speech on your computer in one pass: background noise, hum and fans out, the level evened out, the tone made clear. Light is the default strength, and on Apple silicon Studio mic sound is on, which makes a laptop, webcam or phone mic sound like a studio microphone. Keep background sound leaves the room in for vlogs, and Reduce echo shortens a room's reverb when the studio noise removal isn't doing it already. The cleaned audio is attached to the clip, so the export uses it, and your original audio is kept, so Original audio puts it back at any time. In PandaStudio 2.1.61 and later, a screen recording made with a microphone and System audio keeps the two as separate tracks inside the file. Enhance voice then cleans only your microphone and mixes the computer's sound (a video you play, music, a game) back in untouched, and the panel says Only your microphone was enhanced. Recordings made before 2.1.61, and files where the voice and the background are one track, are cleaned as a whole. For those, mark the parts that should keep their original sound as Original voice: from the start to the end of that region, Enhance voice is off and the clip plays its original audio, in the preview and in the export. Use it for a video you are reacting to, a song, a laugh or a sound effect that the noise removal treats as noise.
In the app
- 1Select the clip and open the Audio tab.
- 2Click Enhance voice. Leave the strength on Light unless you need more; turn on Keep background sound for a vlog.
- 3Turn on Reduce echo if it is offered and the room sounds hollow or boomy.
- 4To hear the difference, press and hold Hold to compare. The button says Playing the original while you hold it, and the enhanced voice comes back when you let go. Holding Space or Enter on the button works too (fixed in PandaStudio 2.1.61).
- 5Click Original audio if you prefer the untouched sound for the whole clip.
- 6To keep the original sound for one part only, such as a video you are reacting to, put the playhead where it starts and click Original voice from here in the Enhance voice panel, or click + on the timeline and choose Original voice here. A 5 second Original voice region appears; drag its ends on the timeline to where the original sound should start and stop (PandaStudio 2.1.61 and later).
- 7To remove an Original voice region, select it on the timeline and delete it.

Ask the AI
Open the AI chat in the editor and say any of these. The agent does the edits for you.
CLI & MCP
Every capability is a verb, reachable from the terminal or from an MCP agent (Claude Code, Cursor, Cline, Continue, Claude Desktop).
pandastudio audio.enhance --id=$ID
Async; poll job.wait. Light by default, Studio mic sound on Apple silicon. --keepBackground=true keeps the room. On a recording with separate mic and system audio, the result says systemAudioKept: true.
pandastudio project.add-original-voice --id=$ID --atMs=80000 --durationMs=45000
PandaStudio 2.1.61 and later. Turns Enhance voice off for that stretch of the edited video and returns a regionId. project.remove-original-voice --regionId removes it.
pandastudio audio.clean --id=$ID --echo=true
Async; poll job.wait for completion. --echo=true also reduces room echo; --echo=false switches back without re-denoising.
pandastudio audio.probe --id=$ID
Check levels and silence before or after cleaning.