# Noise Profile Extractor

Capture a noise fingerprint from a selected region of your audio - floor level, third-octave spectrum, hum harmonics, classification - then reuse it to denoise this file or any other file.

> Canonical page: https://elysiatools.com/en/tools/audio-noise-profile-extractor

- **Category:** Media

- **Keywords:** noise profile, noise fingerprint, noise floor, afftdn, room tone, tape hiss, hum 50hz 60hz, goertzel, spectral flatness, third octave, noise reduction, denoise reuse, batch cleanup

## Overview

Denoising works best when the denoiser knows the noise. This tool captures that knowledge as a reusable fingerprint: mark a region that contains ONLY the unwanted noise (room tone, tape hiss, hum, an air conditioner between sentences), and the extractor decodes it and measures the noise floor level in dBFS, the third-octave band spectrum, the spectral flatness (Wiener entropy - near 1 for white noise, low for tonal hum), and mains-hum harmonics at 50 or 60 Hz plus overtones via exact-frequency Goertzel probes. It then classifies the noise (white hiss, pink, low rumble below 120 Hz, 50/60 Hz hum, broadband, or mixed) and derives tuned FFmpeg afftdn settings from the measurement: the noise floor parameter is set from the measured region level (clamped to the filter’s -80..-20 dB range) and the reduction amount from your chosen strength - rather than leaving both at defaults. Three modes: extract-and-denoise renders a cleaned copy of the whole file through the tuned chain (plus a protective high-pass when hum or rumble dominates); extract-only writes the fingerprint as a JSON file for the record; apply-profile loads a previously saved JSON and denoises ANY file with it - the reuse path for batch cleanup of recordings from the same room, device or tape. The JSON carries the classification, measured floor, band levels, hum frequencies and the recommended filter settings, so the profile is auditable, not a black box. The metadata reports the measured floor before and after denoising (quietest 10 percent of 100 ms blocks), so the effect is verifiable. Feed it honest noise-only regions: a region containing speech will be treated as noise and the denoiser will eat the speech.

## Inputs

- **Audio file** (file)
- **Noise region start (seconds)** (number): e.g. 0
- **Noise region end (seconds)** (number): e.g. 2
- **Mode** (select)
- **Saved noise profile (JSON)** (file)
- **Denoise strength** (select)
- **Output format** (select)

## When to use

- When your recording contains consistent background noise such as room tone, tape hiss, HVAC hum, or electrical buzz.
- When you need an exportable, auditable JSON noise fingerprint documenting floor levels and frequency bands.
- When you want to reuse a single noise profile across multiple audio files recorded in the exact same environment.

## How it works

- Upload your audio file and specify the start and end timestamps of a segment containing exclusively unwanted noise.
- The extractor calculates dBFS floor levels, spectral flatness via Wiener entropy, and 50/60 Hz mains harmonics using Goertzel probes to classify the noise profile.
- The tool configures tuned FFmpeg afftdn parameters based on your selected reduction strength and applies a protective high-pass filter if rumble or hum dominates.
- Download your denoised audio in WAV (24-bit PCM) or lossless FLAC format, or export the noise profile as a JSON configuration file.

## Use cases

- Eliminating constant room ambience or air conditioner hum from voiceover and spoken podcast tracks.
- Digitizing cassette or reel-to-reel tapes by profiling tape hiss from the lead-in pause.
- Batch cleaning interview files captured under identical studio acoustic and microphone setups.

## Frequently asked questions

### What happens if my selected noise region contains speech or vocals?

If speech or vocal content falls inside the noise segment, the extractor treats those vocal frequencies as noise and suppresses them throughout the entire track.

### What information is stored inside the exported JSON noise profile?

The JSON file contains the noise classification, measured floor level in dBFS, third-octave band levels, detected hum harmonics, and tuned filter parameters.

### How does the 'Apply a saved profile' mode work?

It bypasses the extraction step, reading filter settings directly from an uploaded JSON fingerprint to clean a new target audio file without requiring a noise-only section.

### What audio output formats are available for cleaned recordings?

You can export the processed audio as uncompressed 24-bit PCM WAV or lossless FLAC.

### How are the FFmpeg afftdn parameters determined?

The noise floor parameter is derived directly from the measured region level clamped between -80 dB and -20 dB, while reduction depth matches your chosen strength setting.

## Related tools

- [Audio Melody Contour Extractor](https://elysiatools.com/en/tools/audio-melody-contour-extractor): Extract a dominant melody path from audio and export MIDI, note events, a pitch contour, SVG, and JSON in one ZIP.
- [Audio Metadata Remover](https://elysiatools.com/en/tools/audio-metadata-remover): Strip technical metadata from an audio file for privacy before sharing
- [Bass Line Extractor AI](https://elysiatools.com/en/tools/bass-line-extractor-ai): Pull the bass line out of any full mix as isolated audio, a MIDI transcription, or both — powered by local neural stem separation.
- [Image Palette to Design Tokens](https://elysiatools.com/en/tools/image-to-design-tokens): Extract a dominant color palette from an image (k-means clustering), then export it as CSS variables, SCSS variables, a Tailwind config, or JSON design tokens — with named colors and an auto-generated shade scale for each
- [Media Metadata to JSON](https://elysiatools.com/en/tools/media-metadata-to-json): Extract audio/video metadata as JSON with ffprobe.
- [Audio Noise Reducer](https://elysiatools.com/en/tools/audio-noise-reducer): Remove background noise from audio files using various noise reduction algorithms
- [Batch Podcast Publisher Pack](https://elysiatools.com/en/tools/audio-batch-podcast-publisher-pack): Normalize a batch of podcast episodes, apply metadata and chapters, and package validated delivery files in one ZIP.
- [Remix Pack Builder](https://elysiatools.com/en/tools/audio-remix-pack-builder): Turn one music track into a complete remix pack: BPM, key (with Camelot), cue points, four Demucs stems, an MP3 preview, and a metadata JSON — all in one ZIP.

## Samples

- [ELK Stack Log Analysis Samples](https://elysiatools.com/en/samples/elk-stack-samples): Comprehensive ELK Stack (Elasticsearch, Logstash, Kibana) samples for log aggregation, processing, and visualization in distributed systems
- [WebRTC Real-Time Communication Samples](https://elysiatools.com/en/samples/webrtc-samples): Comprehensive WebRTC samples for peer-to-peer audio/video communication, data channels, screen sharing, and signaling server implementation
- [Datadog Monitoring Samples](https://elysiatools.com/en/samples/datadog-samples): Comprehensive Datadog monitoring and observability configuration examples including APM, logs, metrics, and dashboards
- [Express.js Samples](https://elysiatools.com/en/samples/express): Express.js web framework examples including Hello World, routing, middleware, and REST API