Elysia Tools
Navigation
Media
Reduce background noise
Details
Applies ffmpeg frequency-domain denoise (afftdn) or non-local means (anlmdn) to reduce background noise.
Execution
Fill in the form, run the tool, and review the result in one place.
Samples
Related
Ready for a run
Tool usage guide
This tool reduces background noise from audio files using ffmpeg's afftdn or anlmdn algorithms. It allows you to clean up recordings by adjusting parameters like noise floor and strength, and supports various output formats for flexible use.
Background
A voice actor recorded a narration at home, but the microphone captured constant computer fan noise.
Problem
The recording has a low-frequency hum that distracts from the spoken content.
How to use
Upload the audio file, select afftdn mode with a noise floor of -30 dB, and choose FLAC output for high quality.
Outcome
The fan noise is significantly reduced, resulting in a clean and professional narration.
Background
An interview was conducted outdoors where wind and distant traffic added unwanted noise.
Problem
The audio is noisy, making the interviewee's voice hard to hear clearly.
How to use
Use anlmdn mode with a strength of 0.008 to target wind noise, and output in MP3 format for easy distribution.
Outcome
Background noise is minimized, improving the intelligibility and overall quality of the interview.
You can upload audio files in common formats like MP3, WAV, AAC, and others, with a size limit of 200 MB.
afftdn uses frequency-domain denoising for broadband noise, while anlmdn applies non-local means for more targeted noise reduction.
For afftdn, adjust the noise floor in dB; for anlmdn, set the strength parameter to control intensity.
Yes, you can select from formats including MP3, AAC, M4A, OGG, Opus, FLAC, and WAV.
Files must be under 200 MB, and processing time varies based on file size and options, but it typically completes quickly.