Clinic Scribe records a visit in three ways: ambient capture for in-person encounters, system-audio capture for telehealth calls, and dictation when you would rather narrate. All three feed the same pipeline, a live transcript with speaker labels followed by a structured clinical note, so the choice is about how the audio reaches the app, not about a different product for each setting.
| Setting | How Clinic Scribe captures it | Best for |
|---|---|---|
| In person | Ambient capture through your computer's microphone | Office visits, exams, rounds at a desk |
| Telehealth | System audio on your computer, no bot in the call | Video visits on any platform |
| Dictation | You narrate, the scribe structures it | Chart catch-up, quick follow-ups |
In-person visits: ambient capture
For an in-person encounter, Clinic Scribe uses your computer's microphone to listen to the conversation in the room. There is nothing to dictate and no commands to remember: you talk with your patient, and the transcript builds in the background while the app labels who is speaking. Accuracy in this mode depends most on the room, so a low-noise space and a quick mic check make a visible difference. See getting the most accurate transcription.
Telehealth: the call audio, with no bot in the meeting
Telehealth is where the desktop design pays off. Because the patient's voice comes out of your speakers through the meeting app, Clinic Scribe captures the system audio on your computer and transcribes both sides of the conversation. Nothing joins the meeting. Your patient sees you, not a recording participant, which matters for trust and for the awkwardness a visible bot introduces. It works with any telehealth platform, and the app can detect Google Meet and Zoom calls and offer to start a session automatically; see automatic Google Meet and Zoom detection.
On macOS, telehealth capture needs the Screen and System Audio Recording permission. If the patient's voice is missing from the transcript, that permission is almost always the reason; the fix is in fixing recording permissions on macOS.
Dictation: narrate when it suits you
In dictation mode you describe the encounter in your own words and the scribe turns your narration into a structured note. It fits chart catch-up at the end of the day, short follow-ups, and clinicians who already think out loud. You still get a formatted note rather than a raw transcript.
Limits and reliability, stated plainly
- Sessions run up to two hours of net recording (pauses do not count), with a warning before the limit.
- Pause and resume are available at any time, including resuming a session you had ended.
- If your connection drops, the session pauses on its own and protects the transcript, which syncs automatically once you are back online.
- Only one session is active at a time; starting a new one saves and closes the current one.
Frequently asked questions
Does Clinic Scribe join my telehealth call as a bot?
No. For telehealth, Clinic Scribe captures the call audio directly on your computer through system audio. Nothing joins the meeting, so your patient sees only you, not a recording participant. It works with any telehealth platform.
How does Clinic Scribe capture the patient's voice in a video visit?
The patient's voice comes out of your speakers through the meeting app, and Clinic Scribe records that system audio on your computer along with your microphone. On macOS this requires the Screen and System Audio Recording permission, which you enable once in System Settings.
What is dictation mode?
Dictation mode lets you narrate the encounter yourself instead of capturing a live conversation. The scribe turns your narration into a structured clinical note. It suits chart catch-up and quick follow-ups.
How long can a single session be?
Sessions run up to two hours of net recording time, not counting pauses, with a warning before the limit. You can pause and resume at any point, and if your connection drops the session pauses and protects the transcript.