MP3 voice recorder
Record and get an MP3 file, encoded on your own machine. MP3 is the format to pick when the recording has to open somewhere else — an email attachment, a form upload, an old car stereo.
Ready. Press Record and allow microphone access when your browser asks.
Why the browser cannot make MP3 on its own
This is worth knowing, because it explains a delay you may notice. The MediaRecorder API built into browsers produces Opus in a WebM container, or AAC in MP4 on Safari. It has never produced MP3 — no browser vendor implemented it, partly because of the patent situation that persisted until 2017.
So this page captures raw PCM audio from the Web Audio API and runs it through an MP3 encoder compiled to JavaScript, in your tab. When you press Stop you will see a brief "Encoding MP3" message while that runs. A one-minute recording encodes in roughly a second on a modern machine; a thirty-minute one takes noticeably longer. That work is happening on your processor, not ours.
How to record an MP3
- Check that Format is set to MP3 · 128 kbps.
- Press Record and allow microphone access.
- Speak, watching the level meter — blue with amber peaks is the target.
- Press Stop, then wait for encoding to finish.
- Press Download to save the
.mp3file.
Specification
- Container / codec
- MPEG-1 Audio Layer III
- Bitrate
- 128 kbps constant
- Channels
- Mono
- Sample rate
- Matches your device, typically 44.1 or 48 kHz
- File size
- About 0.96 MB per minute
- Encoder
- LAME compiled to JavaScript, running in your tab
- Compatibility
- Universal — every phone, browser, DAW and media player
Is 128 kbps enough?
For speech, comfortably. 128 kbps mono is well above the point where spoken word artefacts become audible, and it is the de facto standard for podcast distribution. Where it starts to matter is music: for singing with instrumental backing, or anything with cymbals and reverb tails, you will hear the compression. Record those in WAV and convert afterwards if you need to.
The other case for WAV is editing. Every MP3 encode discards information permanently, so editing an MP3 and re-exporting it stacks two lossy passes. If the file is going into an editor, start uncompressed.
- WAV recorderUncompressed, for editing and archiving.
- Recording for a podcastLevels, formats and what to do with the file after.
- MP3 vs WAVWhich to pick, with the size and quality trade-offs.
- Audio trimmerTrim the silence off before you send it.
Questions
Why does it say 'Encoding MP3' after I press stop?
Browsers cannot produce MP3 natively, so the page runs an MP3 encoder in JavaScript on your own machine. The pause is that encoder working. Longer recordings take longer, and the file appears as soon as it finishes.
Is the MP3 encoded on my computer or on a server?
On your computer. The encoder is a JavaScript library that loads into the page and runs on your processor. No audio data leaves the tab at any stage.
Can I change the bitrate?
Not currently — the recorder is fixed at 128 kbps mono, which suits speech well and keeps files small. If you need a higher bitrate or stereo, record in WAV and convert with a desktop tool such as Audacity or ffmpeg.
Will the MP3 play on an iPhone?
Yes. MP3 has been supported on iOS since the first iPhone, and the file will play in Files, Music, Voice Memos and any messaging app you attach it to.
The encoding is very slow on my phone. What can I do?
MP3 encoding is processor-intensive and phones are slower at it than laptops. Record shorter takes, or switch Format to the browser's native option — that uses hardware-accelerated encoding and finishes instantly, at the cost of producing a WebM or M4A file rather than MP3.
Does the MP3 have any watermark or attribution audio?
No. The file contains only your recording. There is no audio stamp, no trailing announcement and no metadata tag added.
What sample rate does the MP3 use?
Whatever your audio hardware reports, usually 44,100 or 48,000 Hz. The encoder matches it rather than resampling, which avoids an unnecessary conversion step.