AI Noise Remover SDK · runs in your browser

AI Noise Remover & Loudness Normalizer

Remove background noise from speech, enhance voice recordings and set their loudness to a LUFS target. Your audio never leaves this browser.

Choose an audio file or drop it here · WAV, MP3, M4A, OGG, Opus, FLAC or WebM · up to 10 minutes

Processing Enhance speech Noise removal, rumble filter, gentle compression and loudness normalization Remove noise Neural speech noise removal only Normalize loudness Measure and set loudness, nothing else

Target loudness −16 LUFS −16 for podcasts and voice, −14 for music streaming, −23 for broadcast Noise reduction 100% Lower it to keep some of the room sound

Use it in your own code

This page runs the speech model of the open-source AI Noise Remover SDK with ONNX Runtime Web. The Python and JavaScript packages process complete recordings on your computer or server, with exact two-pass loudness normalization, reports and batch processing.

pip install loudness-audio-kit

from loudness_audio_kit import enhance
enhance("podcast.mp3", "podcast-clean.wav", target_lufs=-16)

Website · Installation guide · GitHub · PyPI · All SDKs and guides

More free audio tools

Audio Enhancer · Echo Remover · Audio Normalizer · LUFS Meter · All Loudness.ai tools · Free Text to Speech in Your Browser

Speech model: DeepFilterNet3 (Hendrik Schroeter and contributors, MIT/Apache-2.0), streaming export by deepfilter-stream (MIT). See NOTICE and the third-party folder in this Space. Processing quality on music or heavily reverberant rooms is limited; the model is made for speech. Example recording: Garth Comira reading Ashiel Mystery, chapter 2 (LibriSpeech, CC BY 4.0), with added noise.

Choose audio

Result

Use it in your own code

More free audio tools