No bot in your call
The app records locally from your microphone. No participant appears in the meeting, and nothing enters the call on WhisperJot's behalf.
WhisperJot turns what your microphone can hear into a live transcript with timestamps — ideal for in-person meetings, interviews, lectures, and standups in a room. Meeting transcription runs on-device by default with the local engines on macOS, Windows, and Linux (beta).
Start the recording, watch the transcript take shape, and leave the note-taking to WhisperJot.
What can leave your Mac: Identify Speakers (optional) uploads the meeting's retained audio for speaker labels. The separate invite-only meeting-agent beta runs in the cloud. Optional cloud library sync uploads transcript text, titles, and timestamps, never audio.
Product behavior checked against WhisperJot's meeting implementation on August 24, 2026. Meetings and dictation share the loaded engine, so they do not run simultaneously.
The app records locally from your microphone. No participant appears in the meeting, and nothing enters the call on WhisperJot's behalf.
The desktop microphone workflow uses local engines by default. Optional Identify Speakers uploads retained meeting audio for speaker labels after you choose it.
After a one-time model download (about 500 MB on Mac, around 600 MB on Windows and Linux), meeting transcription works without an internet connection.
Transcript segments persist during the meeting. After an interruption, WhisperJot finalizes the meeting at next launch and retains its audio.
Find past meeting transcripts locally in History. On macOS, the menu bar also shows live meeting status.
Export Markdown (.md) or plain text (.txt), copy timestamped lines to the clipboard, or reveal the WAV recording on disk.
The full audio recording is retained on your machine; optional Identify Speakers uploads it for speaker labels. Audio retention is configurable and defaults to 90 days; transcripts are never auto-deleted. Optional cloud library sync is opt-in and syncs text, titles, and timestamps, never audio.
Use it for in-person meetings, interviews, lectures, standups in a room, and your side of calls. Meeting notes are included in the flat $99/year plan, with no per-minute caps or storage tiers and unlimited meetings within the six-hour per-meeting cap.
The desktop microphone workflow does not capture the other side's system audio. It produces a transcript without AI summaries; optional Identify Speakers adds speaker labels. What you get is a private, timestamped transcript you can search and export.
The desktop microphone workflow records locally, so nothing joins the call as a participant. It is designed for in-person meetings, interviews, lectures, standups in a room, and your side of a call. The separate invite-only meeting-agent beta runs in the cloud.
Meeting transcription runs on-device by default with the local engines. The full audio recording is retained on your machine; optional Identify Speakers uploads it for speaker labels. Optional cloud library sync is opt-in and syncs text, titles, and timestamps, never audio.
Yes. After a one-time model download (about 500 MB on Mac, around 600 MB on Windows and Linux), meeting transcription works without an internet connection.
Each meeting has a six-hour safety cap. Meetings and dictation share the loaded transcription engine, so they do not run at the same time.
Yes. Export the timestamped transcript as Markdown or plain text, or copy timestamped lines to the clipboard. You can also reveal the meeting's WAV audio file on disk.
No. WhisperJot produces a timestamped transcript that you can search locally and export. Summaries are not a current feature.
Explore meeting transcription for consultants and journalists, compare WhisperJot, or review the $99/year plan. See the best meeting transcription software roundup and Granola vs Otter.
Live, timestamped meeting transcription on your own device — with no bot on the guest list.