Record your voice online
Record your microphone with a live waveform, pause and resume, and save WebM, MP3 or WAV. No time limit, no account, nothing uploaded.
- Runs in your browser
- Files are never uploaded
- Free, no account
Your browser will ask for microphone permission. Nothing is captured until you press record, and the permission applies to this site only.
Microphone settings
Device names appear after you allow microphone access once.
Removes sound coming back in from your speakers. Turn it off when recording music or an instrument.
Cuts steady background noise like fans and air conditioning. It can dull quiet speech and thin out singing.
Levels out volume as you move nearer and further from the mic. Off gives a more faithful, less processed recording.
How to record
- Press record and allow microphone access when the browser asks. The permission is granted to this site only, and nothing is captured before you press the button.
- Watch the waveform and the level bar. Speech should push the bar to somewhere between 40% and 80% — a bar that barely moves means the gain is too low, one that pins at 100% means clipping.
- Pause and resume as often as you like; the segments join seamlessly.
- Stop, listen back in the player, then download as-is or convert to MP3 or WAV.
Where the recording lives while you work
The browser’s MediaRecorder hands back a chunk of encoded audio once a second, and
those chunks are held in an array in this tab’s memory. When you press stop they are assembled
into a single Blob and given to the player. Nothing is written to disk until you press download,
and nothing is sent anywhere at any point — there is no server behind this page to send it to,
and the site’s Content-Security-Policy lists no third-party origin.
The practical consequence: reloading or closing the tab destroys the recording. There is no draft saved in the cloud, which is the trade-off that comes with there being no cloud.
What format you get, and whether to convert
| Browser | Container and codec | Roughly |
|---|---|---|
| Chrome, Edge, Firefox, Opera | WebM containing Opus | 30 MB per hour |
| Safari (macOS and iOS) | MP4 containing AAC | 45 MB per hour |
Both are complete, playable files that need no further processing. Convert only when the destination demands it:
- MP3 for anything going to older software, a car stereo, a voicemail system, or a transcription service with a narrow decoder.
- WAV for a DAW or sampler, where uncompressed PCM avoids repeated decoding — see MP3 to WAV for what that costs in size.
- Neither if you are simply keeping the file or sending it to someone on a modern device. WebM and MP4 audio play in every current browser and phone.
The three processing switches
Browsers apply voice-call signal processing to microphone streams by default. All three are exposed in the advanced panel because the right answer depends entirely on what you are recording.
| Setting | Leave on for | Turn off for |
|---|---|---|
| Echo cancellation | Anything recorded through laptop speakers, or with audio playing | Music, instruments — it can duck sustained notes as if they were echo |
| Noise suppression | Speech in a room with fans, traffic or air conditioning | Singing, acoustic instruments, ambience you want to keep, very quiet speech |
| Automatic gain control | Interviews where people move around, or when levels vary | Performances with intentional dynamics; it flattens loud and quiet alike |
Getting a better recording
- Distance beats equipment. A cheap headset mic 10 cm from your mouth will beat an expensive one across the room, because room reflections are what make recordings sound amateur.
- Speak past the microphone, not into it. Positioning it slightly off-axis avoids the plosive blast on p and b sounds.
- Kill the reflective surfaces. A recording made facing a bookshelf or a curtain sounds dramatically better than one facing a bare wall or a window.
- Test first. The microphone test checks levels and the play-back loop in a few seconds, before you commit to a long take.
- Record a few seconds of silence at the start. Noise-reduction tools in editors use it as a profile.
Is there really no time limit?
There is no imposed one, because there is no per-minute cost to anybody. The real constraint is memory: at roughly 30 MB an hour for Opus, a multi-hour session is entirely realistic on a desktop and less so on an older phone. For anything over about two hours, stopping and starting a new take periodically is sensible insurance against a browser crash — and you can stitch the takes back together afterwards with the audio joiner, losslessly if they are all the same format.
What this cannot do
- Record system audio.
getUserMediaonly exposes microphone inputs. Capturing what is playing on your computer needs screen-capture permission with audio sharing, and on macOS an additional loopback driver. - Record multiple microphones separately. One input device at a time; a multitrack session needs a DAW and an audio interface.
- Transcribe. That requires a speech model, which means either a large download or a server. Neither fits here, and we would rather say so than pretend.
- Survive a reload. Nothing is stored, by design.
Questions
Is my recording uploaded anywhere?
No. The browser hands the microphone stream to MediaRecorder inside this tab and the chunks are held in memory. There is no server behind this page and no request is made while you record — open your Network tab, or switch off Wi-Fi, and record anyway to prove it.
Is there a time limit?
None is imposed. Recording length is bounded only by memory: Opus at the default quality is roughly 30 MB per hour, so multi-hour sessions are realistic on a desktop. For very long recordings, stopping and starting a fresh take every so often is still wise insurance against a browser crash.
What format do I get, and should I convert it?
Chrome, Edge and Firefox record WebM containing Opus; Safari records MP4 containing AAC. Both are finished, playable files. Convert to MP3 if you are handing the file to someone using older software, or to WAV if it is going into a DAW — each runs one extra local FFmpeg pass.
Should I leave noise suppression on?
For speech, yes — it removes fan and air-conditioning noise convincingly. Turn it, echo cancellation and automatic gain all off when recording singing or an instrument: the algorithms are tuned for voice calls and will duck sustained notes and flatten dynamics.
Can I record what is playing on my computer?
Not from here. getUserMedia only exposes microphone inputs. Capturing system audio requires screen-capture permission with an audio share, which is what a screen recorder does; on macOS it additionally needs a loopback driver.