Converter directory
Select your video format to see the available audio outputs. Every converter runs in your browser, so your file stays on your device.
A video file is a container holding at least two streams: pictures and sound. Extracting audio means discarding the video stream — which is never decoded — and re-encoding what is left. That is why conversion speed tracks the length of the soundtrack rather than the resolution or file size of the picture, and why a 4K movie often finishes as quickly as a 720p clip of the same duration.
Two engines do that work here. Formats the browser already understands — MP4, M4V, MOV, QT, MKV and WebM — stream through the native WebCodecs pipeline with nothing to download and no practical size ceiling. Everything else — AVI, FLV, WMV, 3GP, MPEG and transport streams — goes through FFmpeg compiled to WebAssembly, which fetches about 30 MB on first use and decodes inside a 32-bit memory space. The result is the same; the setup cost and the size limit are not.
Before you pick a page below, decide what the audio is for. That one choice matters more than any other setting, and it is the difference between a 30 MB file and a 700 MB one.
Five exports appear across these pages. None of them is universally right — each wins in a specific situation.
Listening on any device
The most compatible audio format ever shipped — every car stereo, phone, speaker and operating system since the late 1990s plays it. Choose 128 kbps for speech, 192 kbps as the general default, 320 kbps for music you intend to listen to on decent headphones.
Size: 0.96 MB per minute at 128 kbps
Editing, transcription, restoration
Uncompressed 16-bit PCM. Nothing is lost and every editor accepts it, which is why it is the correct handoff for a mix, a transcript, or a noise-reduction pass. The price is size — roughly 10.6 MB per minute at 44.1 kHz stereo.
Size: 10.6 MB per minute (44.1 kHz stereo)
Apple devices and smaller files
AAC is roughly 20-30% more efficient than MP3, so a 192 kbps M4A is comparable to a 256 kbps MP3. It is the native format of iPhone, iPad, Music, GarageBand and Logic. Older car stereos and cheap speakers may not play it.
Size: 0.96 MB per minute at 128 kbps
Archives and music libraries
Lossless compression at roughly 55% of WAV's size, with a checksum so corruption is detectable. Every sample comes back identical on decode. Apple's own apps do not play FLAC directly, so treat it as storage rather than delivery.
Size: About 5.8 MB per minute (44.1 kHz stereo)
Open pipelines, Wikimedia, game engines
Patent-free and fully open. Required by Wikimedia Commons and treated as a first-class import by engines such as Godot. Hardware support is patchy and Apple's ecosystem ignores it entirely — pick it because something downstream needs it.
Size: 0.96 MB per minute at 128 kbps
Uses the browser's own WebCodecs pipeline. Nothing to download, no meaningful size ceiling, and the file is read progressively rather than loaded into memory all at once.
Handles the legacy formats no browser decodes. Downloads about 30 MB once per page load and works inside a 32-bit memory space, so keep these files under roughly 500 MB.
1 audio option
2 audio options
1 audio option
1 audio option
2 audio options
5 audio options
3 audio options
1 audio option
2 audio options
2 audio options
MP4, M4V, MOV, QT, MKV, WebM, AVI, FLV, WMV, ASF, 3GP, 3G2, MPG, MPEG, MTS, M2TS and TS. The first six run on the browser's native engine; the rest are handled by FFmpeg compiled to WebAssembly.
Formats the browser can decode natively — MP4, M4V, MOV, QT, MKV and WebM — stream straight through the WebCodecs pipeline with nothing to download. AVI, FLV, WMV, 3GP and MPEG require FFmpeg WebAssembly, which downloads about 30 MB on first use and runs in a 32-bit memory space.
We impose no cap. Native formats are streamed, so a multi-gigabyte MP4 or MKV works. The FFmpeg WebAssembly path is limited by its 32-bit memory space: comfortable under about 500 MB, likely to fail above roughly 2 GB. For large legacy files, trim them first or extract the audio with desktop ffmpeg.
No. The file is read through a BlobSource inside your tab and the audio comes out of that same tab. There is no backend to upload to — disconnect from the network after the page loads and the conversion still completes. Open DevTools and watch the Network panel if you want to verify.
MP3 at 128 kbps for speech-only episodes, or M4A at 192 kbps if you prefer a smaller file with equivalent quality. Normalise loudness to about -16 LUFS in your editor before publishing — this converter does not change gain.
Those are cinema formats browsers do not decode. Convert the audio locally first with ffmpeg -i input.mkv -c:a aac -b:a 192k -vn audio.m4a, then drop that much smaller file in here. This is the single most common fix for a failed conversion.
No. Video bought from Apple is wrapped in FairPlay DRM, which no browser can decode and which we do not attempt to bypass. Files you exported yourself — from iMovie, HandBrake, QuickTime Player or a screen recording — work normally.
Yes. Every output is a fresh encode at the bitrate you choose, not a copy of the source stream. If you need bit-identical audio, export WAV or FLAC — both are lossless.
No. Tags, artwork, chapters and subtitles live outside the audio stream and are dropped on export. Add them afterward in your tagger or podcast host.
Still unsure which page you need? Start with MP4 to MP3, the most common combination, or read the about page for how VideoToAudioConverter is built and maintained.