Skip to content

Use OpenWhisper

Meeting mode

Capture a call, configure optional AI insights, and export the results.

On this page

Meeting Mode captures microphone and system audio, builds a transcript, and optionally generates AI notes and insights. Windows and macOS support meeting capture; Linux system audio remains a preview.

Prepare audio capture#

Choose your microphone and a speech engine that supports meetings. On macOS, allow Microphone and Screen & System Audio Recording in Privacy & Security.

On Linux, system audio needs native PulseAudio or PipeWire with the pipewire-pulse compatibility service. The app checks the default output's monitor source. Follow the Linux system-audio guide if readiness checks fail.

Run a short test with audio playing before your first full meeting. Confirm that your microphone and the other participants are both present in the recording.

Choose a speech engine#

In Settings → Meeting Mode → Voice & speakers, choose a supported speech engine. Qwen3-ASR and cloud OpenAI transcription are dictation/upload options in this release; use an engine with meeting support from Choosing a model.

For a paired host, choose Remote computer here. Audio is sent to the host for transcription. See Remote engines for setup and storage options.

Add optional AI intelligence#

Pi is the default meeting engine. Install it from Downloads → Components, then select your text endpoint and credentials in the meeting intelligence settings. The packaged OpenCode SDK is another available component.

You can also choose an already installed and signed-in Claude Code, Codex, or OpenCode agent in Settings → Meeting Mode → Intelligence. OpenWhisper detects these installations; it does not install, update, or sign into them for you. Their meeting passes use that agent's account or API usage.

If the selected AI engine is unavailable, install or update it, select another, or continue recording without AI insights. Recording remains available without AI intelligence.

Record and review#

Open Meeting Mode, add an optional brief, and start the recording. The brief guides AI notes and the live copilot and can be edited during the meeting.

After transcription, Continue in the background lets you start another meeting while reports finish. Find saved results in Past Meetings. Review important names, decisions, and commitments against the transcript before sharing a report.

Export a meeting#

Use the meeting dashboard or Past Meetings export dialog. Markdown can include the transcript and AI insights; plain text is a transcript format; JSON retains the structured exported state.

Exports include recording references where applicable, not the audio files. For format details and the full-meeting download behavior, see Export support.

Checked against OpenWhisper 2.6.14.App documentation
Still stuck?

Get help from the community, or tell us what went wrong.