Meeting transcription
A readable meeting transcript with timestamps.
Upload a recording or use your browser microphone, then review the local transcript before creating notes.
In short: Meeting Recorder transcribes MP3, WAV, M4A, MP4, microphone recordings, and permitted browser audio with Whisper in the browser. The first run downloads and caches the local model, then produces timestamped lines with editable speaker labels. In local mode, the audio is not sent to a transcription API; a separate AI summary request sends transcript text to the configured summary provider.
Open the browser recorder Turn the transcript into AI notes
How to use it
- 01 — Choose a recording source
Start microphone or browser-audio capture with participant consent, or upload a file you are allowed to process.
- 02 — Run and review transcription
Let local Whisper create timestamped lines, then check wording, timestamps, and speaker labels.
- 03 — Export or create notes
Keep the transcript, request optional AI notes, and export the version you reviewed as Markdown, TXT, JSON, or PDF.
Example meeting notes
Research interview excerpt
Summary: The participant described onboarding friction and agreed to review a revised flow after the next prototype test.
Decisions: Test the revised onboarding flow · Review findings after the next prototype
Action items: Unassigned — share the prototype link · Interviewer — add the quote to the research notes
Frequently asked questions
Which file types are supported?
The upload workflow accepts MP3, WAV, M4A, and MP4, in addition to recordings made in the browser.
Is the audio uploaded to a transcription service?
Not in local mode. Whisper runs in the browser after its model is downloaded, so the audio is not sent to a transcription API. A custom remote API mode can be configured separately and has its own disclosure.
Are speakers identified automatically?
The transcript provides editable labels such as Speaker 1. Local Whisper is not presented as a guaranteed speaker-diarization system; review or rename labels before sharing.
Can I jump to a timestamp?
Yes. Timestamp controls in the transcript can move the audio player to the corresponding point when an audio preview is available.
Can I make AI meeting notes from the transcript?
Yes, when the summary provider is configured. The transcript remains the source, and only transcript text is sent when you explicitly request the AI summary.
Record or upload the source
Use a browser microphone or permitted tab/window capture, or upload MP3, WAV, M4A, or MP4 from a call, interview, class, or research session.
Run Whisper locally
The first run can take longer because the browser downloads and caches the local Whisper model. The default local path keeps the audio on this device.
Timestamps, with honest speaker boundaries
Transcript lines include timestamps and an editable speaker label. The local workflow does not promise automatic diarization, so review labels when multiple people are speaking.
Create notes after review
Use the transcript as the source for a summary, decisions, and action items. If you request AI notes, only transcript text is sent to the configured summary provider.