1 · Recording quality: 8 kHz vs 16 kHz
Both clients record at either 8 kHz (smaller files) or 16 kHz (clearer audio, roughly double the file size). The defaults differ by client — the Chrome extension defaults to 8 kHz (a "🎚 HQ" toggle switches it to 16 kHz), while the mobile app defaults to 16 kHz and automatically steps down to a smaller upload if a file would be too large for the connection.
2 · Templates, with examples
A template is a Name and a Body — the report skeleton for a study type you dictate often. The name always acts as a spoken trigger; add extra trigger phrases under Advanced. Templates sync across all three clients: build one on the phone, use it in the extension or desktop the same day.
Applying one is entirely by voice — there's no menu to pick from. While dictating, say the template's name (or one of its Advanced triggers) and the AI recognizes it from the list of your templates, then builds the report on that skeleton instead of transcribing from a blank page; anything else you say gets woven into the template body.
Example. Template named "Normal Chest", body = a standard normal chest X-ray report, Advanced trigger "CXR normal". Saying "normal chest" or "CXR normal" at the start of a dictation invokes it.
The edge case worth knowing about. If you dictate something onto a template that doesn't map onto any part of it — a stray measurement, an incidental finding — the AI won't guess where to put it. It's pulled out and shown in a persistent toast with a COPY button instead, so it's never silently dropped; you paste it into the report by hand. See the worked examples in the Mobile and Extension guides.
3 · Editing your report
On mobile, the finished report sits in an editable text box right in the app — tap in and fix anything; the edit saves automatically the moment you tap away. On the extension and desktop, there's no separate "edit report" screen — the report text lands directly in your RIS/HIS field, so you edit it the same way you'd edit anything else typed there.
4 · Copy & Transfer
The extension's History tab (and the mobile Reports tab) can hand a past report back to you two ways: COPY puts the full text on the clipboard to paste anywhere; ➡ TRANSFER (extension only) inserts it into the RIS exactly like a live dictation — same cursor-target and findings/conclusion split rule covered in §7. Click into the target box first, then transfer. Full detail: extension guide §6.
5 · Dictate on phone
When the workstation's microphone is bad (or nonexistent), press 📱 DICTATE ON PHONE in the extension or desktop app. Your signed-in phone pops up a record prompt; you dictate there, and the finished report travels back end-to-end encrypted and inserts into the browser exactly like a live dictation — same insertion rule as everything else on this page. Full walkthroughs: mobile guide §9, extension guide §4.
6 · No audio lost to a bad connection
Two mechanisms, one goal — a dropped connection never costs you a dictation:
- Mobile's offline queue. A dictation that can't reach the server is force-saved and queued on the phone, then sent automatically (retrying on its own) once you're back online, the app reopens, or it returns to the foreground. Repeated failures surface as FAILED so you can retry by hand — nothing is dropped quietly.
- Retranscribe, on both clients. Every backed-up recording (mobile Audio tab, extension REC tab) has a ↻ Retranscribe button that re-runs the saved audio through transcription from scratch and saves a fresh report. Use it after a failed transcription or any time you want a second pass.
The honest difference: mobile auto-retries on its own; the extension doesn't have a background queue — a failed live dictation instead offers RETRY/SAVE on the on-page pill mid-flow, and REC-tab Retranscribe is the manual redo afterward. Either way, the audio itself is never silently lost.
7 · Cursor position & the findings/conclusion split
The rule that governs where text lands, everywhere it's inserted (live dictation, Upload Audio, phone dictation, and History Transfer alike):
- Text always lands where your cursor was when the dictation started — click into the box you want before you begin.
- On reporting systems with separate findings/conclusion fields (e.g. APEX-style forms), the report is split automatically between the two boxes — but only if both boxes are empty at the moment dictation starts.
- If either box already has text in it (a loaded template, notes you typed), the split is skipped and the whole dictation is inserted as one block into the box your cursor is in — nothing existing is overwritten.
Something this guide doesn't answer? Contact us — or use the feedback form.