Recording and transcription

DocNote records the encounter on your phone or in the browser, then transcribes it with medical-aware speech-to-text.

1. Open the Record tab
2. Choose a template
3. Tap record and speak
4. Pause or stop when the encounter is done

You can add more audio to the same consultation later from the report screen (Add recording).

On the web app, use a recent Chrome, Edge, Firefox, or Safari on https. If the browser blocks the microphone, recording will not start.

A consultation recording on the web app can last up to 3 hours. At 3 hours DocNote stops it, saves it, and shows a message. To keep going, use Add recording on the same consultation.

Radiologists work in a split screen, whatever template they use: the report on the left, your dictation as it is transcribed on the right. Your specialty decides, in Settings → Account: choose Radiology (split screen) for this view, or Radiology (single screen) to keep the classic recorder — record, then generate the report — with the same radiology templates. Other specialties use the classic recorder; the generic Dictation template shows the transcription live above it.

• Exam title with the date sits under the report name. It stays on: each integration writes the exam title and the consultation date on the first line. Turn it off to leave that line out. A title already in the template, or one you dictate, stays.
• Spoken commands stay as words in the live transcription (virgule, point, à la ligne). Integration turns them into a comma, a period, or a line break, and drops the command words. One à la ligne is a single line break, with the next words directly under the previous line. Two in a row leave a blank line and start a paragraph, and so does paragraphe (or nouveau paragraphe).
• Click in the report to choose where the next dictation goes. Until you click, each dictation is placed in the matching section. After a click, the dictation goes exactly where you clicked (the end of the conclusion included) and the cursor moves behind the inserted text, so a correction dictated in several takes stays together. “The next dictation goes to the cursor” shows under the report; press its ✕ or Esc to release it.
• A dictation that starts with a section name (“Conclusion…”, “Technique…”) always goes into that section, wherever the cursor is.
• If you pause for more than about 20 seconds before dictating again, the cursor is released and the next dictation is placed in its section again — click again to choose a spot.
• Pause the microphone and the dictation is folded into the report automatically. There is no button to press.
• What an integration has just written appears in blue text in the report, so you can check what was added and where without rereading everything. The blue is on screen only — it never reaches the copy, Word or your RIS — and it clears as soon as you type.
• You can type in the report while DocNote integrates your dictation: your corrections are kept.
• In the assistant, Use as note adds the answer to the live transcription as if you had dictated it; it is then integrated into the report like any dictation.
• The live transcription scrolls with you, so the words you are saying stay in view.
• Use the eye in the transcription header to hide that panel; it hides itself once you validate the report, and comes back as soon as you record again.
• The formatting bar (bold, italic, headings, tables) stays out of the way until you press the Modify pencil at the top right of the report.
• The consultation appears in the list on the left as soon as recording starts.
• Leaving the screen keeps that consultation, including a draft you have not validated. It stays in the list until you delete it yourself.

Transcription starts automatically after the audio is uploaded.

Status	Meaning
Pending	Queued for processing
Processing	Actively transcribing
Processed	Ready to use for generation
Failed	An error occurred — you can retry

Transcription understands drug names, dosages, and abbreviations, and works in 20+ languages. You can dictate in one language and generate the report in another from your consultation-language setting.

Once audio is transcribed (and any attachments are processed), generate or regenerate the report. New audio added while a report is already generating is picked up on the next pass.

See Record a consultation.