Voice Recorder - Record Audio Online

How to use this tool

1. Click "Start Recording" and allow microphone access.
2. Speak. The timer shows elapsed recording time.
3. Click "Stop" when finished.
4. Preview the audio then click "Download" to save the file.

Audio format support

The recorder checks WebM with Opus, WebM, OGG with Opus, OGG, MP4 with AAC, then MP4, using the first MediaRecorder type your browser reports as supported. Safari can record MP4/AAC when supported; OGG is selected only when the browser reports it. The format label and downloaded extension follow the selected browser-native recording container.

Common audio formats and quality

This is a common-format comparison, not a list of export options. Voice Recorder downloads only the selected browser-native recording container; it does not create WAV or MP3 files. Actual codec support and recording quality depend on the browser and microphone.

Opus at 32–64 kbps can deliver clear speech at small file sizes, although the browser chooses the actual bitrate. WAV is useful for uncompressed editing, while AAC in an MP4/M4A container is common in Apple software and podcasts. Those formats are included here for comparison; this page downloads only the browser-native container selected for the take.

Privacy

Audio capture, encoding, and preview happen in the page; recorded chunks stay in memory until you download them to your device. The page does not upload recordings. The optional Remember control stores your microphone choice in this browser so the same input can be selected next time.

MediaRecorder API limitations

The Web MediaRecorder API has several constraints worth knowing:

• No pause control: this tool lets you stop a take, but does not pause and resume an in-progress recording.
• Browser-controlled bitrate: while a target bitrate can be requested, browsers may not honour it exactly.
• Memory limits for long sessions: recording for many hours continuously accumulates data in memory. For long recordings, stop and save periodically.

Post-recording workflow

• Transcription: upload to OpenAI Whisper, Otter.ai, or similar services for automatic speech-to-text.
• Cleanup: use Audacity (free, cross-platform) or Adobe Podcast Enhance (browser-based) to reduce background noise and improve clarity.
• Format conversion: FFmpeg (ffmpeg -i input.webm -c:a libopus output.ogg) converts to any format. HandBrake provides a GUI alternative.
• Distribution: upload to podcast hosts (Buzzsprout, Spotify for Podcasters, Transistor) which generate RSS feeds and distribute automatically.

Sample rate guidance

Human speech is intelligible up to about 8 kHz, so a 44.1 kHz sample rate is overkill for voice recordings. By Nyquist’s theorem, a sample rate of 2× the highest frequency is sufficient - 16 kHz captures everything the human voice produces, and 22.05 kHz provides a comfortable margin. Lower sample rates produce smaller files with no perceptible quality loss for voice content.