Strip background noise out of any speech recording with the RNNoise model running entirely in your browser. Fans, traffic, room hum and HVAC drone disappear; the voice stays intact.
AI Noise Reducer is under active development. We're building browser-based audio tools powered by FFmpeg WebAssembly. Check back soon!
AI Noise Reducer is shaped around how people actually use audio editing and conversion utilities online: open the page, drop in a file, get the result. Strip background noise out of any speech recording with the RNNoise model running entirely in your browser. Fans, traffic, room hum and HVAC drone disappear; the voice stays intact. The interface stays out of the way once the work begins so the engine can use the available CPU and memory for the actual transformation.
AI Noise Reducer runs on the RNNoise speech-denoising model compiled to WebAssembly — an open-source, well-audited engine that performs the audio editing and conversion natively in the browser. It accepts MP3, WAV, OGG, FLAC, M4A, and AAC and produces output that opens in any standard audio viewer. Per-run input is capped at 200 MB.
Common audiences for AI Noise Reducer include language learners reviewing speech and sound designers prototyping cues, but plenty of people land on the page through a one-off search and never come back — that is also fine. The tool is built to be useful even when you only ever need it once.
Because everything runs in the page, the tool scales the same way for one user or a million — there is no per-user backend cost. The page is static, the engine is the same JavaScript bundle for every visitor, and the work happens on the visitor's own device. That keeps the tool free and keeps it fast on the first interaction.
The right moment to reach for AI Noise Reducer is when you have a focused audio editing and conversion job that fits inside a browser tab. Open the page, drop in the file or paste your input, choose the options that matter, and the tool returns the result.
Once the engine finishes, `{name}-denoised.wav` is offered as an immediate download. There is no preview gate, no email-wall, and no "register to download" intermediary — the file is yours the moment it is ready.
The architecture imposes only the limits the browser itself imposes. The published 200 MB ceiling is conservative; most modern devices comfortably handle inputs up to that size, and the cap exists so the tool degrades gracefully on phones and budget laptops rather than running out of memory.
Even on its own, AI Noise Reducer composes well with the rest of your toolkit. The output is a standard MP3 file that opens in any program that handles the format, so the result of one run can become the input to whatever step you use next.
The transformation in AI Noise Reducer is deterministic — the same input plus the same options produces the same result every run. That predictability matters when the result has to match an upstream specification or be reproducible later.
AI Noise Reducer is structured around the idea that a useful tool should be its own page. Open the page, do the work, close the tab — the page is the entire product. There is no onboarding flow because there is nothing to onboard into.
AI Noise Reducer runs as a regular web page, so there is no install step or permission grant before the first run. The page can be audited by viewing the source or by watching the developer-tools Network tab while a job runs.
Useful patterns when working with AI Noise Reducer: keep the input file open in another tab so you can compare against the result; give the output file a descriptive name when saving so you can find it later (the default name is sensible but generic); and treat each run as independent — the tool has no concept of "history", which means you cannot accidentally pollute one job with leftovers from another.
If AI Noise Reducer appears to hang, the engine is almost certainly still working — large inputs simply take longer to process inside a browser than they would on a server with multi-core scheduling. For inputs near the 200 MB cap, give it up to a minute on a typical laptop before assuming something is stuck.
That is the whole tool. Use AI Noise Reducer for as long as it stays useful to you, and if it does, the catalog has many more tools built the same way. Each applies the same single-purpose discipline, so the way you used this page transfers to the next one you try.
Stationary background noise: HVAC, fans, traffic, room hum, hard-drive whine. The model is trained on speech so it leaves the voice untouched.
Not at the default strength. RNNoise applies a per-frame gain that suppresses noise without re-synthesizing speech. Aggressive settings can introduce mild artefacts on consonants.
A noise gate cuts off the entire signal below a volume threshold — fine for between-word silence but it kills quiet speech too. RNNoise predicts noise per frame and subtracts only the noise component.
No. RNNoise compiled to WebAssembly runs locally; the model bytes (~14 MB) download once and cache in your browser.
WAV is lossless so the cleaned audio is not subjected to a second lossy encoder pass. Convert to MP3 / AAC afterwards with Audio Converter if size matters.
AI Noise Reducer is built to preserve quality wherever the underlying audio format allows it. Operations that are mathematically lossless (e.g. structural transformations, lossless re-encoding) round-trip with no perceptible change. Operations that involve a lossy codec inevitably introduce small artefacts at the byte level, but the defaults aim at the sweet spot where output looks or sounds the same to a normal viewer or listener while still being meaningfully smaller or faster than the input.
AI Noise Reducer works in any modern browser released in the last few years — Chrome, Edge, Firefox, Safari, Brave, Arc and the major Chromium derivatives are all supported. The underlying engine relies on widely-supported web APIs, so there is nothing exotic to install. If you are on a very old browser version and the tool fails to load, updating to the latest release of your preferred browser is the only fix needed.
Favtoo keeps no copy of your file because Favtoo never receives your file. AI Noise Reducer runs entirely in your browser, the input is held only in your tab's memory, and closing the tab discards it. There is no opt-in cloud history, no "recent jobs" panel synced to an account, and no server-side retention to configure — the architecture simply has nowhere for your file to be stored.
AI Noise Reducer processes one input at a time by design — it keeps memory usage predictable on lower-end devices and makes results easier to verify. To handle a folder, run the tool once per file; the page stays loaded between runs and remembers your last-used settings, so the second run is essentially instant.
AI Noise Reducer is free to use. The processing runs in your browser, which keeps the per-user cost low enough that the tool can be offered openly. The download is the same file the engine produced — you can use it for as many runs as you need.
Inputs are capped at 200 MB per file, which keeps memory usage stable across phones, tablets and older laptops. You can run AI Noise Reducer as often as you need; every run produces a full-quality result.
No installation is needed. AI Noise Reducer runs as a normal web page, with no browser extension, no native helper, and no separate desktop client to download. That is partly a privacy choice — extensions can request broad permissions, while a regular page is sandboxed by default — and partly a convenience one: you can use AI Noise Reducer on any computer you have temporary access to without leaving anything installed on it.
Most jobs finish in seconds. Speed scales with input size and with how many CPU cycles your browser tab has available — the engine runs in your browser, so it shares resources with whatever else you have open. For inputs near the 200 MB ceiling, expect anywhere from a few seconds to roughly a minute on a typical laptop. Closing other heavy tabs noticeably speeds things up.
AI Noise Reducer is updated whenever the underlying engine releases an improvement or a bug fix. Because the tool is delivered as a static page, every visit fetches the latest version automatically — there is no "version" to manage on your end. If a particular release ever changes default behaviour, the change is documented on Favtoo's changelog so you can confirm what shifted.
Compress Audio
Shrink any audio file to a smaller size by lowering the bitrate. Pick a target quality (96, 128, 192, 256, or 320 Kbps) or output format (MP3, OGG, M4A) and the file is re-encoded right inside your browser using FFmpeg WebAssembly. Nothing is uploaded — your audio never leaves your device.
Convert Audio
Convert any audio file between MP3, WAV, OGG, FLAC, M4A, AAC, and Opus right in your browser. Pick the output format and (for lossy formats) the target bitrate. Everything runs locally with FFmpeg WebAssembly — your file is never uploaded and no account is required.
Audio Recorder
Record from your microphone directly in the browser. Pick quality (high, medium, low), toggle echo cancellation, noise suppression and auto-gain, then save to WebM/Opus or M4A/AAC. Audio is captured locally — nothing is uploaded.
Text to Speech
Type or paste text, pick a system voice, and listen instantly. Adjust speaking rate (0.5×–2×), pitch, and volume in real time. Uses your browser's built-in Web Speech API — no cloud TTS, no API keys, no costs.
Tone Generator
Generate a pure tone at any frequency from 20 Hz to 20 kHz. Pick a sine, square, triangle, or sawtooth waveform, choose duration, amplitude, and mono/stereo. Exports a 16-bit PCM WAV file at 44.1 kHz with built-in click-preventing fades.
Silence Generator
Generate a perfectly silent WAV file of any length from 1 second up to 1 hour. Pick mono or stereo, get a 16-bit PCM WAV at 44.1 kHz. Useful as padding between clips, intro silence, leader audio for video timing, or test material.
White Noise Generator
Generate white, pink, or brown noise as a 16-bit PCM WAV file. Pick noise type, duration up to 1 hour, amplitude, and mono/stereo. Useful for sleep, focus, masking distractions, audio testing, and as a backing layer for ambient music.
Metronome
A precise browser-based metronome powered by the Web Audio API. Set BPM from 30 to 300, choose a time signature, accent the first beat, and use tap-tempo to sync. Click timing is sample-accurate using lookahead scheduling — much steadier than typical JavaScript setInterval beats.
