Is the recording uploaded anywhere?

No. Sound from the microphone stays inside the browser, and the only way it leaves is the download you start yourself. The decoding and encoding done for the export also happen on this page rather than on a server. Recording does require microphone permission, and it is the browser that asks for it — all this tool can do is receive audio from a microphone you have already allowed.

Should I save WAV or MP3?

WAV to hand to an editor or a transcription service, MP3 for sharing and storage. WAV is uncompressed — it discards nothing, and costs roughly 16 MB a minute in stereo (this tool writes 24-bit; a 16-bit file would be two thirds of that). MP3 at 192 kbps is about 1.4 MB a minute, and 320 kbps about 2.4 MB. For spoken material 192 kbps is plenty; going higher buys very little you can hear. If you are unsure, keep the WAV — you can always make it smaller later.

My recording disappeared after I stopped

The take exists only inside the page, so a reload, a closed tab or a restart all lose it. There is no autosave: leaving recordings of someone’s voice lying around on their device without being asked is not a good default for a tool like this. Nothing is kept until you press Download, so save as soon as you have checked the take, and save in sections during a long session.

Can it remove noise or raise the volume?

This tool records and nothing more. Level and tone belong to the mastering tool, separating vocals from backing to the stem separator, loudness measurement to the key & BPM analyser, key and tempo to the key & tempo changer, and format conversion to the file converter. The intended path is to record here, save as WAV, and take that file to whichever page does the next job.

Does it work on a phone?

Yes, with two caveats. Another app holding the microphone — a call, a voice memo — stops a take from starting. And switching the screen off or moving the browser to the background can end a recording, because the operating system decides to. For a long take on a phone, keep the screen on and this page in front. If you are using a headset microphone, connect it before opening the page.

The browser owns the microphone permission

The prompt comes from the browser itself, not from this page. That is why choosing Block once makes every later attempt fail silently — from the outside it looks like a button that does nothing. Open the padlock (or camera/microphone icon) in the address bar, set this site back to Allow, and reload the page. On a phone there is a second layer: the browser app needs microphone permission from the operating system as well. If the site is set to Allow and recording still fails, check the system settings too.

Recording needs an https connection

Browsers only grant microphone access over an encrypted connection, or on localhost during development. The rule exists so that no page delivered over a tappable connection can open your microphone. This site is always served over https, so it works here — but the same HTML saved to disk and opened as a file, or hosted on an internal server without encryption, cannot record. When that is the situation, the page says so instead of failing quietly.

Speech-oriented processing is on by default

Browsers normally hand over audio with echo cancellation, noise suppression and automatic gain already applied. That suits calls and voice memos. For an instrument it softens the attack, pumps the level of quiet passages up and down, and thins out the sound of the room. "Record without processing" asks for all three to be switched off — but it is a request, and some devices and browsers start recording without honouring it. Listen back and check whether the level of the quiet passages stays put. The setting can only be changed before a take begins.

Long recordings use the device’s memory

Audio accumulates inside the page until it is written out. Tens of minutes are usually fine; several hours can make a modest device sluggish, and in the worst case the tab ends and the take goes with it. For long sessions, record in parts and save as you go. The recording also exists only inside the tab, so reloading the page or closing the window discards it — nothing reaches disk until you press Download.

Back to all tools／ Music & audio

Voice recorder

Recording needs microphone permission, which your browser will ask for. The audio itself stays inside this page and is never uploaded.

Record straight in the browser. Watch the input level as you go, pause and resume, then play the take back and save it as WAV or MP3. No app, no account.

Options

CheckingRecorder

0:00.0

Input level

−∞

Result

Your browser will ask for permission to use the microphone. Recording starts once you allow it.

The input level is visible while you record, which heads off the most common failure of all: discovering on playback that the microphone was picking up nothing.

All tools

Highlights

You can see the level as you record: A meter runs the whole time. The two commonest ways a recording goes wrong — ten minutes of talking into a muted microphone, and clipping you only discover on playback — both come down to nothing on screen showing the signal. Reaching 0 dB, and recording with no signal arriving at all, are each called out in words as well as on the meter.
Pause and pick up where you left off: Pause for a thought or an interruption and resume into the same take. Everything up to the point you stop lands in a single file, so there are no fragments to line up and stitch together afterwards. Nothing is captured while paused, so the break does not appear in the recording.
Saved in a format everything opens: Browsers do not agree on what to record into: Chrome and Firefox use WebM, Safari uses MP4. Plenty of players and editors will not open a WebM audio file at all. The export reads the take back and rewrites it as WAV or MP3, so it is ready to hand to a transcription service or an editor as it is.

How to use it

Press Start and allow the microphone
The browser asks for permission to use the microphone; choose Allow. Recording begins and the elapsed time and input level start moving. For instruments or room sound, tick "Record without processing" before you start.
Record, watching the level
Peaks around −12 dB are a good target. Use Pause to break off and Resume to carry on in the same take. Press Stop when you are finished.
Listen, then save
Stopping reveals a player so you can check the take. Choose WAV or MP3 (MP3 offers a bitrate as well) and press Download to keep the file. "Record again" starts a fresh take.

Before you record

The browser owns the microphone permission: The prompt comes from the browser itself, not from this page. That is why choosing Block once makes every later attempt fail silently — from the outside it looks like a button that does nothing. Open the padlock (or camera/microphone icon) in the address bar, set this site back to Allow, and reload the page. On a phone there is a second layer: the browser app needs microphone permission from the operating system as well. If the site is set to Allow and recording still fails, check the system settings too.
Recording needs an https connection: Browsers only grant microphone access over an encrypted connection, or on localhost during development. The rule exists so that no page delivered over a tappable connection can open your microphone. This site is always served over https, so it works here — but the same HTML saved to disk and opened as a file, or hosted on an internal server without encryption, cannot record. When that is the situation, the page says so instead of failing quietly.
Speech-oriented processing is on by default: Browsers normally hand over audio with echo cancellation, noise suppression and automatic gain already applied. That suits calls and voice memos. For an instrument it softens the attack, pumps the level of quiet passages up and down, and thins out the sound of the room. "Record without processing" asks for all three to be switched off — but it is a request, and some devices and browsers start recording without honouring it. Listen back and check whether the level of the quiet passages stays put. The setting can only be changed before a take begins.
Long recordings use the device’s memory: Audio accumulates inside the page until it is written out. Tens of minutes are usually fine; several hours can make a modest device sluggish, and in the worst case the tab ends and the take goes with it. For long sessions, record in parts and save as you go. The recording also exists only inside the tab, so reloading the page or closing the window discards it — nothing reaches disk until you press Download.

Questions

Is the recording uploaded anywhere?: No. Sound from the microphone stays inside the browser, and the only way it leaves is the download you start yourself. The decoding and encoding done for the export also happen on this page rather than on a server. Recording does require microphone permission, and it is the browser that asks for it — all this tool can do is receive audio from a microphone you have already allowed.
Should I save WAV or MP3?: WAV to hand to an editor or a transcription service, MP3 for sharing and storage. WAV is uncompressed — it discards nothing, and costs roughly 16 MB a minute in stereo (this tool writes 24-bit; a 16-bit file would be two thirds of that). MP3 at 192 kbps is about 1.4 MB a minute, and 320 kbps about 2.4 MB. For spoken material 192 kbps is plenty; going higher buys very little you can hear. If you are unsure, keep the WAV — you can always make it smaller later.
My recording disappeared after I stopped: The take exists only inside the page, so a reload, a closed tab or a restart all lose it. There is no autosave: leaving recordings of someone’s voice lying around on their device without being asked is not a good default for a tool like this. Nothing is kept until you press Download, so save as soon as you have checked the take, and save in sections during a long session.
Can it remove noise or raise the volume?: This tool records and nothing more. Level and tone belong to the mastering tool, separating vocals from backing to the stem separator, loudness measurement to the key & BPM analyser, key and tempo to the key & tempo changer, and format conversion to the file converter. The intended path is to record here, save as WAV, and take that file to whichever page does the next job.
Does it work on a phone?: Yes, with two caveats. Another app holding the microphone — a call, a voice memo — stops a take from starting. And switching the screen off or moving the browser to the background can end a recording, because the operating system decides to. For a long take on a phone, keep the screen on and this page in front. If you are using a headset microphone, connect it before opening the page.

Options

Result

Related tools

Highlights

How to use it

Press Start and allow the microphone

Record, watching the level

Listen, then save

Before you record

Questions