Online Voice Recorder screenshot

Online Voice Recorder

Record from your microphone in the tab, with an explicit permission request, a live level and a download of the in-memory recording. Nothing is uploaded.

Privacy First
0 Runs
Updated 2026-11-22
Permission on press
Stream stays in the tab
No upload
In-memory Blob

What this page makes

Opens the microphone on your press, records with MediaRecorder, shows a timer and level, and downloads the Blob the browser produced.

How to use it

Press Start recording and allow the permission prompt. The timer and level run while the recorder is open. Press Stop recording, play the result in the page, and press Download recording to save the one file the browser encoded.

Worked example: a 12.400 s spoken sentence with the default microphone becomes one in-memory WebM/Opus Blob, a 12.400 s timer reading, and a download button for that file.

Method

Start calls getUserMedia({audio: true}) only after your press, so the permission prompt is the first thing that happens. The stream feeds a MediaRecorder, whose first supported container the page names: WebM/Opus, OGG/Opus or MP4. Stop assembles the chunks into one in-memory Blob and stops every track. A live analyser reports the input level in dBFS.

What it refuses

If the browser exposes no MediaRecorder or no getUserMedia, the page names the missing feature instead of offering a dead button. A denied permission, a missing device and a page served over plain http are each reported as the named state they are.

Where it runs

The microphone opens only on your press, and the audio never leaves the tab: there is no upload, no server call and no storage entry. Closing the page discards an undownloaded recording. The browser requirement is stated on the page: Recording needs getUserMedia and MediaRecorder. A page served over plain http is granted no microphone, and a browser without either API is told so by name.

The rest of the family

The other audio and recurrence pages are listed under Calculators: tone and binaural generation, the virtual piano, pitch shifting, looping and reversal, the stated noise filter chain, mid/side cancellation, microphone recording and the closed-form recurrence solver.

Opens the microphone on your press, records with MediaRecorder, shows a timer and level, and downloads the Blob the browser produced.

Start calls getUserMedia({audio: true}) only after your press, so the permission prompt is the first thing that happens. The stream feeds a MediaRecorder, whose first supported container the page names: WebM/Opus, OGG/Opus or MP4. Stop assembles the chunks into one in-memory Blob and stops every track. A live analyser reports the input level in dBFS. The microphone opens only on your press, and the audio never leaves the tab: there is no upload, no server call and no storage entry. Closing the page discards an undownloaded recording.

Worked example: a 12.400 s spoken sentence with the default microphone becomes one in-memory WebM/Opus Blob, a 12.400 s timer reading, and a download button for that file.

The timer and level come from the live stream, and the download is exactly the Blob the browser encoded.

How to use the online voice recorder

Press Start recording and allow the permission prompt. The timer and level run while the recorder is open. Press Stop recording, play the result in the page, and press Download recording to save the one file the browser encoded.

The microphone opens only on your press, and the audio never leaves the tab: there is no upload, no server call and no storage entry. Closing the page discards an undownloaded recording.

  • Domain: One microphone, whatever container the browser encodes, downloaded as one file.

What the page proves

Opens the microphone on your press, records with MediaRecorder, shows a timer and level, and downloads the Blob the browser produced.

Start calls getUserMedia({audio: true}) only after your press, so the permission prompt is the first thing that happens. The stream feeds a MediaRecorder, whose first supported container the page names: WebM/Opus, OGG/Opus or MP4. Stop assembles the chunks into one in-memory Blob and stops every track. A live analyser reports the input level in dBFS.

  • Browser requirement: Recording needs getUserMedia and MediaRecorder. A page served over plain http is granted no microphone, and a browser without either API is told so by name.
  • The page states its domain: One microphone, whatever container the browser encodes, downloaded as one file.

What it refuses, and what it leaves out

If the browser exposes no MediaRecorder or no getUserMedia, the page names the missing feature instead of offering a dead button. A denied permission, a missing device and a page served over plain http are each reported as the named state they are.

  • Not the right page for: multi-track recording
  • Not the right page for: a recording that survives a reload
  • Not the right page for: choosing a codec or a bitrate
  • Reported as a named state: a denied permission is a named state, not a page failure
  • Reported as a named state: a page served over http is granted no microphone
  • Reported as a named state: a browser without MediaRecorder cannot encode a recording

What to Expect

Best for

  • recording a voice memo without an account
  • checking a microphone before a call
  • a quick take you will download immediately

Not ideal for

  • multi-track recording
  • a recording that survives a reload
  • choosing a codec or a bitrate

What this tool keeps

  • exactly the bytes the browser's MediaRecorder produced

What may need cleanup

  • download before closing the tab, because an undownloaded Blob is discarded
  • use headphones to avoid feeding the speakers back into the microphone

Common errors

  • a denied permission is a named state, not a page failure
  • a page served over http is granted no microphone
  • a browser without MediaRecorder cannot encode a recording

Example use cases

  • record a voice memo
  • check which microphone the browser picked
  • capture a quick idea and download it

Sample input

a 12.400 s spoken sentence with the default microphone

Sample output

one in-memory WebM/Opus Blob, a 12.400 s timer reading, and a download button for that file

Who this is for

  • anyone recording a quick voice note
  • readers checking a microphone setup

Frequently asked questions

Is my data sent anywhere?

The microphone opens only on your press, and the audio never leaves the tab: there is no upload, no server call and no storage entry. Closing the page discards an undownloaded recording.

Which browser API does this page need?

Recording needs getUserMedia and MediaRecorder. A page served over plain http is granted no microphone, and a browser without either API is told so by name.

What can I download?

Download recording saves the one Blob the browser's MediaRecorder produced.

Where is the recording stored?

In this tab's memory as a Blob. It is not written to storage and not uploaded. Closing the page discards it, so download it first.

What format is the download?

Whichever container the browser's MediaRecorder supports first, named on the page: usually WebM with Opus, sometimes OGG/Opus or MP4. The page does not transcode.

Can I record system audio or another tab?

No. getUserMedia opens a microphone input. Capturing a tab or the system needs a screen-capture API and a different permission, which this page does not request.

Developer program

Build custom tools for your store

Join the SDK program to create useful ecommerce utilities for your store.