WEBM to MP3 converter

What is actually inside a WEBM file

WEBM is a container, not a codec. It is a restricted form of Matroska, published by the Google-sponsored WebM Project in 2010 as an open, royalty-free format for the web, and it holds one or more compressed streams side by side. The audio stream is either Opus or Vorbis. Opus is what you will find in almost anything recorded recently; Vorbis is its predecessor and still turns up in older files. There is often a video stream too, compressed with VP8 or VP9, or AV1 in some newer files.

The extension tells you none of this. A file called recording.webm might be a two-minute voice note or a forty-minute screen capture with a soundtrack, and you find out which by opening it. For this converter the difference barely matters: it takes the audio stream, whatever codec it uses, and leaves everything else behind.

Where WEBM files come from

A lot of WEBM files were never chosen as a format by anyone. A web page recorded them, and WEBM is simply what the browser wrote.

• Anything recorded in a browser. When a site records your microphone, camera or screen, it usually does it with the browser’s built-in MediaRecorder API, and in Chrome and Edge that produces WEBM with Opus audio unless the site asks for something else. Online voice recorders, browser-based screen recorders and extensions, and web tools for recording interviews or podcast guests hand you .webm files for exactly this reason. Firefox can write WEBM as well, although for audio on its own it defaults to Ogg, a different container around the same Opus audio. Safari recorded only MP4 until version 18.4, released in March 2025, added WEBM.
• Some downloads. Several sites serve their audio as Opus in a WEBM container, so a file you saved expecting an MP3 can arrive as .webm instead.

If your file came from an iPhone or a Mac’s own apps instead, it is far more likely to be M4A, and the M4A to MP3 converter is set up for those. Uncompressed recordings from an audio interface or a field recorder are usually WAV, which is what the WAV to MP3 converter is for.

How to convert WEBM to MP3 on this page

1. Add your files. Drag them onto the drop zone or click to pick them; you can add several at once. The file picker is set to show WEBM files, but anything else your browser can decode is still converted — you will just see a note that it is not a WEBM.
2. Choose the output. Bitrate is 128, 192, 256 or 320 kbps, with 192 as the default. Channels is either Keep original or Mono.
3. Press Convert all. Files are converted one at a time. Each row shows the file name, its size and duration, its status, a progress bar while it encodes and, when it is done, the size of the MP3.
4. Download. Every finished row has its own Download button, and Download all saves each MP3 as a separate file. Your browser may ask once whether the page is allowed to download multiple files.

Cancel stops the file that is encoding at that moment and halts the rest of the batch; Convert all picks up again from there. Remove takes a single row off the list, and Clear empties it. Each MP3 is named after its original, with spaces and unusual characters turned into hyphens: Team call (final).webm comes out as Team-call-final.mp3.

What happens to your file, and what never does

Every step runs inside the browser tab you are looking at. Here it is in order, because “nothing is uploaded” is a claim you should be able to check.

• Reading. The file is read off your disk into the tab’s memory. That step does not involve the network at all.
• Decoding. The Web Audio API’s decodeAudioData hands the bytes to the same decoder your browser uses to play a WEBM video, and gets uncompressed samples back at the rate your audio hardware runs at, usually 44.1 or 48 kHz. The MP3 keeps that rate; a higher one, such as 96 kHz, is brought down to 48 kHz, the highest rate MP3 supports. Only the audio track is decoded; video frames are never touched.
• Encoding. The samples go to LAME, the long-established open-source MP3 encoder, compiled to WebAssembly and running in a Web Worker — a background thread of the same tab, which is why the page stays responsive and the progress bar keeps moving while a long file encodes.
• Saving. The finished MP3 exists only in memory until you click Download, and then it is saved by your browser like any other download.

Files go through that pipeline one after another rather than all at once, because a decoded file takes far more memory than the WEBM it came from. If you want to verify the privacy claim, open your browser’s developer tools, switch to the Network tab and convert a file: the only requests are for the site’s own code, the MP3 encoder’s scripts among them once you press Convert all, and not a byte of your audio goes anywhere.

Choosing a bitrate for audio that was already compressed

Converting WEBM to MP3 is a lossy-to-lossy conversion. Opus already discarded detail when the file was made, and no MP3 setting brings it back; the only goal is to avoid losing more. That makes the choice simpler than it looks:

• Speech — voice notes, meetings, interviews, lectures: 128 kbps is plenty, and Mono makes sense for one voice, because the whole bitrate then goes on a single channel instead of being shared between two.
• Music or anything mixed: 192 or 256 kbps, with the channels kept as they were.
• Not sure: leave it at 192 kbps, the default.

320 kbps is the maximum. Because every setting is a constant bitrate, it makes each file a quarter larger than 256 does, and like every other setting it cannot put back what the original encoding removed.

Do not be surprised if the MP3 is larger than the WEBM you started with. Opus is a much newer codec than MP3 and reaches the same quality at a lower bitrate, so matching it with MP3 takes more bytes. That is the price of compatibility, not a fault in the conversion. Mono averages the channels together; a recording that was mono to begin with stays mono whichever option you choose.

Screen recordings and other WEBM video

A WEBM screen recording converts exactly like an audio file, and the MP3 contains its soundtrack alone. The video is dropped from the output, and the original file on your disk is not changed. If the recording was made without sound — a screen share with audio switched off, say — there is no audio track inside it, nothing to convert, and the row will report that the file could not be decoded.

If what you actually want from a recording is the words rather than the sound, the audio to text converter transcribes a recording, WEBM included, with Whisper running on your own device, and the MP4 to text tool does the same for video files.

Browser recordings with no duration

A WEBM saved by Chrome’s MediaRecorder often has no duration and no seek index. The recorder writes the start of the file while recording is still under way, when the length is not yet known, and never goes back to fill it in. The symptom is a player that shows no running time or will not let you scrub, and it is a long-standing known limitation of how browsers write these files, not damage.

It does not stop the conversion. The browser decodes the audio track from start to finish, so even when a row starts with no duration, the real figure appears as soon as that file has been decoded. The MP3 this page writes is constant bitrate: every second takes the same number of bytes, so a player can work out its length and seek positions from the file size alone.

WEBM in Safari, and on iPhone

Chrome, Edge and Firefox decode WEBM audio in Opus or Vorbis without trouble, and they are the browsers to use here. Safari came to WEBM late. Safari 14.1 on the Mac added WEBM with VP8 or VP9 video and Vorbis audio in April 2021, Safari 15 added Opus audio in WEBM that September, and Safari 17.4 listed VP8, VP9 and WEBM support for iPhone and iPad in March 2024. Decoding a WEBM file’s audio through Safari’s Web Audio API is still unreliable, and when it fails the row says so in plain words rather than showing a generic error.

On an iPhone or iPad the other browsers have generally been built on the same engine as Safari, so switching browser there is unlikely to help. A computer running Chrome, Edge or Firefox is the dependable route.

Why MP3, when WEBM already plays in a browser

WEBM is at home in a browser and much less so everywhere else. Apple Podcasts accepts MP3 or AAC audio in a podcast feed, not WEBM. Audacity opens WEBM only after you install its optional FFmpeg library. QuickTime Player on a Mac does not open it at all. MP3, by contrast, is the format that nearly every player, editor, car stereo and upload form accepts without question, which is why it is still the safe choice for a file that has to work wherever it ends up.

One thing the conversion does not carry over is metadata. The converter writes audio only: the MP3 has no ID3 tags and no cover art, and any title stored inside the WEBM is left behind. Add tags afterwards in your music library or a tag editor if you need them.

Limits worth knowing before you start

• 200 MB per file, with a warning above 50 MB. Decoding expands the audio into uncompressed samples in memory, and an hour of stereo at 48 kHz comes to about 1.4 GB once decoded, however compact the WEBM was.
• Audio only, from video as well as audio files.
• No tags or artwork in the output.
• No editing. The whole file is converted. To cut out a section first, use the audio trimmer; to stitch several recordings into one, the audio joiner.

Where VoiceSnap Pro fits

A lot of the WEBM files that get converted to MP3 are someone talking: a voice note recorded in a browser tab, an answer to an interview question, a thought captured before it disappeared. Often the next step is typing it up. VoiceSnap Pro is built to skip that step. It is a voice-to-text dictation app for macOS and Windows: hold one keyboard shortcut, speak, and clean punctuated text appears in whatever field your cursor is already in — an email, a Slack message, a document. Every dictation is also saved to a searchable notes library. It is a one-time $39 purchase, not a subscription.

It has not shipped yet, so there is nothing to download today. Join the waitlist for a single email on release day. In the meantime, the voice typing tool shows what dictated text reads like in your browser, and the audio to text converter will transcribe the recordings you already have.