Voice Recorder Studio
A little space for your words
Download HTML
Your browser. Your voice.

From spoken to written.

Speak and see your words appear. Edit, replay and listen back.

1 Record→2 Transcribe→3 Listen

English transcription runs on this device. First use downloads the speech model. Completed audio and text are also saved to Creative studios in Google Drive when connected.

Browser + Google Drive copies · connect Creative studios.

Transcript
Ready to transcribe
00:00
Recording time

Choose Start & transcribe, wait for Listening, then speak.

Live words appear here as you speak. You can edit them after stopping.
0 wordsDraft on this device
Saved in VRS on this browser
0 items
Your saved recordings and notes will appear here.

Timed captions

Recording preserves recognised phrase times. Review timing and text before sharing. Editing the transcript does not rewrite these cues.

Live audio chunks are saved in this browser. Completed outputs also appear in Creative studios, with a Google Drive cloud copy when connected. Clearing browser data removes browser copies; check the Creative studios save status and keep important downloads.

Made for listening, too.

Use Read aloud in the bottom-right corner. Listen to the whole transcript, your selection, or from the cursor. Collapse the player to keep reading while listening.

Where are my recordings and texts saved?

Live recording, playback and English recognition run on this device. Completed audio and nonempty transcripts are saved in the shared browser output library as well as the VRS session list. Open Creative studios and connect Google Drive to save copies in your Creative studios cloud folder for exchange with the other studios and devices. Pending copies wait until you connect. The app reports a Drive copy as saved only when the shared library confirms the cloud save. The optional browser speech service may process live audio online. Keep Local voices only checked for local read aloud.

Why is transcription unavailable?

Choose Audio + Transcription, the Device transcriber and an English language, then choose Start & transcribe. First use prepares the included model before opening the microphone. Wait for Listening and transcribing before speaking. Words appear in the main transcript while you talk and can change until confirmed. Other languages need a supported native language pack. If the local engine cannot start, the app explains the problem before recording; it does not silently record without the requested transcription.

What are the recorder and input device?

The recorder runs in this browser on your device. The input device supplies the sound: choose your built-in microphone, a headset microphone or a USB microphone. The Device transcriber receives that same input for English. On Linux, an input named “Monitor of…” usually captures computer playback rather than your voice. Browser speech service and native language packs may use the system default microphone separately.

Opening the downloaded HTML

Use the live HTTPS app for immediate automatic transcription. The downloaded HTML contains the interface and app logic; the included English recognizer also needs its companion assets. Download the complete source and build package from the project folder and follow its README to serve the complete app on localhost. Drive preview does not run microphone applications. Keep the app tab open while recording.

Recovery and limitations

The app checkpoints audio chunks and transcript changes on this device. After an interrupted session, it offers the saved portion. Browser crashes or storage limits can lose recent chunks; some interrupted audio containers cannot be played. Recovery is best effort. Read-aloud progress follows text, not the original audio timestamps. Browser synthesis does not provide an audio file to export.

Audio recorderLive transcriptionRead aloudOn-device engine

Confirm