Dictating

In short. Dictation is turned into text on your phone, and your audio stays there. Say the structure first, read the words before you tap Capture, and correct anything wrong. If the phone keeps mishearing you, there is an optional cloud mode, off unless you turn it on.

Dictating well

A workable dictation:

“TKR right knee for severe OA in a 65-year-old man. Tourniquet 90 minutes. Cemented implant. Primary surgeon. Slightly difficult exposure due to obesity.”

Tap the microphone next to Notes. The words appear as you speak. Tap again to stop.
Say the structure first: “right knee TKR” before the long description. If the tail is lost, the important fields still land.
Say numbers as words: “ninety minutes” works better than “9-0 minutes”.
Read the words before you tap Capture. The app cleans up common medical terms and learns from your edits, so a term you correct once is more likely to come out right next time.
Operative times are picked up. Say “started 9:15, finished 10:45” or “took about ninety minutes” and the agent fills in the operative time. Earnings are never dictated.

You do not need to speak up or hold the phone close; the app evens out the loudness of your recording on the phone.

Setting up voice happens the first time you tap the microphone, if you skipped it during sign-up: the app asks for the microphone permission, and may download a one-time voice pack. That first recording starts a little slower. Settings offers a Make dictation instant prompt to do it ahead of time.

Higher-accuracy cloud transcription

Transcription on your phone is free, works offline, and the audio never leaves the device. It cannot be told your accent or your sub-specialty vocabulary in advance. If it keeps mishearing the same terms, you can switch on cloud transcription.

Where to turn it on. Open Settings and switch on High-accuracy cloud dictation. It applies to every microphone in the app, on this device only; turn it on again on another phone or browser. If the switch is greyed out with a note that it isn’t available yet, the option has not been enabled for your account. The voice screen shows the same switch before you record.

What is sent. The first time you turn it on, the app shows a short consent, because this changes what leaves your phone:

Off unless you turn it on. Until then, everything stays on your device.
The audio of that recording is sent, including any patient name you say aloud, to a secure medical transcription service that works under a signed healthcare data agreement. If that matters for a case, leave it off and dictate without the name, or type the notes.
Nothing is kept. The audio and transcript are deleted straight away, and nothing is stored on our servers.
You always know which mode you are in. A small High-accuracy cloud badge sits beside the words while you record. No badge means nothing left your phone.
The words that come back are checked the same way. Patient name and MRN are still removed on your phone before anything is saved, and you still read the words before you tap Capture.

You can switch it off again at any time.

Less common

Three seconds of silence stops the recording. Tap the microphone to carry on. A second dictation is always added to what is there; it never replaces it. On the voice screen, Record more continues from your words. In a notes field the new words go in where your cursor was.
Three minutes per recording. Long enough for a full dictation.
Every microphone is the same. Notes on an existing case, the feedback form and manual capture all open the same recording view.
If the cloud connection drops, the app keeps recording. The badge shows that words will appear when you finish, and the whole recording is transcribed when you stop. If the cloud cannot be reached at all, the app uses your phone instead. You never lose a dictation.
Where the rest of your data goes is on Where your data lives.