Zoom Recording Cleaner for Mac: Echo, Noise, Filler Words
Updated: May 2026
Zoom recording file formats and what you are actually working with
Zoom saves local recordings in one of two formats depending on your settings. Audio-only recordings save as M4A (AAC audio in an MPEG-4 container). If you have "Record to cloud" enabled or use certain export settings, you may get MP4 video files with embedded audio tracks. Aulio Studio reads both formats directly -- you do not need to convert or extract audio first.
One common point of confusion: Zoom's "separate audio" recording option saves each participant on a separate M4A track. This gives you more control in post but means you have multiple files to process. Aulio Studio handles this with batch processing: queue all participant tracks, apply the same preset, and process them all in one run.
Multi-speaker recordings and compounding noise
In a Zoom recording with multiple participants, each participant's background environment contributes to the mixed file. If three participants are on a call, you may have HVAC from participant A, traffic from participant B, and keyboard noise from participant C, all mixed into the same audio stream. This is more challenging than single-source noise because the noise sources do not share a consistent spectral profile.
Neural noise reduction handles this better than profile-based approaches precisely because it does not rely on a static noise sample. The model separates voice from non-voice patterns across the full spectrogram, which means it responds to each noise source independently rather than trying to match a single estimated noise floor. The result is clean speech from all three participants without having to process their tracks separately.
Zoom's own audio processing: friend and foe
Zoom applies real-time noise suppression and echo cancellation during the call, before the audio is recorded. This is generally helpful -- it reduces the live noise during the conversation. The recorded version is already partially cleaned. The downside: Zoom's suppression can introduce its own artifacts (pumping, residual noise, slightly processed quality) that interact with post-processing in ways that natural room noise does not.
For best results, apply a moderate noise attenuation setting in Aulio Studio rather than an aggressive one. The recording may already have had its noise floor reduced by Zoom's processing, so you may only need 15 to 20 dB of additional attenuation rather than the 25 to 35 dB typical for unprocessed recordings. Use the live preview to find the right level for your specific recording.
Filler words in meeting and interview recordings
Meeting recordings -- especially interviews and podcast conversations conducted over Zoom -- often have higher filler word density than in-person recordings. The technical difficulty of conversing over a latency-affected connection encourages more hedging, more um-and-uh filler, and more false starts as speakers navigate interruptions and talk-overs. If you are repurposing a Zoom interview as podcast content, filler removal is worth the extra review step.
Aulio Studio detects fillers in Zoom recordings the same way it does in any audio: the on-device speech model timestamps each one and presents them for review. You see each filler in context and decide whether to cut it. For an interview, this typically surfaces 10 to 30 cuts in a 30-minute conversation -- a manageable review that meaningfully improves the listener experience.
Getting a clean transcript from a Zoom recording
Zoom generates its own transcripts for cloud recordings, but the accuracy depends heavily on audio quality and is often poor for noisy or multi-speaker recordings. Aulio Studio's on-device transcription runs on the cleaned audio -- after noise removal has been applied -- which significantly improves accuracy. Running transcription after noise reduction rather than on the raw Zoom file is the single biggest improvement you can make to transcript quality.
Once the transcript is generated, you can export it as TXT, or use it for clip export: select a quote or segment in the transcript and export just that audio clip. Useful for social media snippets, highlights, or show notes quotes without any waveform scrubbing.
Aulio Studio home · Pricing · For podcasters · Alternatives · Support