Tota is a Mac dictation app that transcribes clinic letters, notes and referrals on the Mac itself, ships with a medical starter glossary, and never sends patient audio or text to a server.
Clinic overran, there are eleven letters to dictate, and the hospital's transcription backlog is three weeks. You could type them. You could use a cloud dictation service and hope the information-governance team never asks. Or you could press a shortcut on the Mac in the clinic room, speak the letter, and have it appear in the letter template with the drug names spelled correctly, with nothing leaving the machine.
Why patient data and cloud dictation are a governance problem
Clinical dictation has always been private by accident: a tape or a digital recorder, a secretary, a letter. Cloud dictation apps changed that by sending the audio to a vendor's servers, where it is transcribed and sometimes passed through a third-party AI provider. Each hop is a disclosure of identifiable patient information that your information-governance process has to sign off.
The vendor answer is a certification and a contract: HIPAA in the United States, data-processing agreements under UK and EU law, SOC 2 reports. Those control what happens to the data after it has been shared. They do not remove the sharing.
Tota removes the sharing. Whisper speech recognition, the local formatting model and speaker identification all run on the Mac. There is no account, no upload and no Tota server that ever sees patient content, and the app works in a basement clinic with no signal.
What clinicians actually dictate
- Clinic letters. The letter to the GP after an outpatient appointment, dictated into your letter template or the free-text field of the EHR.
- Referrals and discharge summaries. Structured documents where Smart Formatting turns 'new paragraph' and 'new list' into layout.
- Ward and consultation notes. Short entries typed directly into the record's free-text field.
- Medico-legal reports and research notes. Long, careful documents that are much faster to talk than to type.
- Transcripts of recorded consultations. Where consent has been given, Audio Drop transcribes the recording on-device with speaker labels.
What leaves your Mac, and what doesn't
This is the whole list. Every row is checkable from the app's own behaviour, and the security page explains how to watch its network activity yourself.
| Data | Leaves your Mac? | What actually happens |
|---|---|---|
| Your voice while dictating patient notes | No | Recorded to a temporary file, transcribed on your Mac by Whisper via WhisperKit, then deleted. |
| The transcript | No | Typed into the app you are using. Kept in a local history for three days, then archived as Markdown in a folder you choose. Entries can be deleted. |
| Text on your screen | No | With “Spell what’s on your screen” on, the visible text of the window you are dictating into is read once, to bias spelling. Never stored or sent. Can be turned off. |
| Glossary terms and learned corrections | No | Stored as JSON files in Tota’s folder in Application Support. |
| Smart Formatting | No | Runs a local Qwen 3 language model via MLX. Off by default. |
| Audio files dropped in for transcription | No | Transcribed and speaker-labelled on your Mac. |
| Licence key and a machine identifier | Yes | Sent to api.lemonsqueezy.com at activation and validation. Never any content. |
| Update check | Yes | A download request to github.com at launch. Can be disabled in Settings. |
| Model downloads | Yes, one-way | Speech, formatting and speaker-ID model files from huggingface.co, once, on first use. |
| Usage analytics or crash reports | Never | There is no analytics or crash-reporting SDK in the app. |
How Tota works for doctors and clinicians
Install the app, grant it microphone and accessibility access, and press Cmd+Shift+R wherever your cursor is. Tota records until you press it again, transcribes the audio on your Mac with OpenAI's Whisper model running through WhisperKit, and types the result into the field. There is also a hold-to-talk key for short entries and a “Hey Tota” wake word for hands-free use. It behaves like a keyboard, so it works in browser-based systems, Word, email, Notes and anything else with a text field, with no integration to request from IT. After the one-time model download it needs no internet at all.
Three optional layers sit on top. Smart Formatting runs a small local language model that fixes punctuation, turns “new paragraph” and “new list” into structure, and corrects stutters without rewriting what you meant. The auto-learning dictionary remembers the corrections you make. And Audio Drop transcribes audio files you already have, with automatic speaker labels, also on-device.
Workflows for a clinic
- The clinic letter. Cursor in the letter body, shortcut, speak. Drug names and conditions come through correctly because the Medical glossary is biasing the model while it transcribes.
- Notes into a browser-based record. Tota types into any text field, including web EHRs and practice systems. There is no plugin to get approved; it behaves like a keyboard.
- The twenty-second ward note. Hold-to-talk for a one-line entry, then back to the patient. Filler-word removal keeps it clean.
- Offline on purpose. Many clinical rooms have no usable Wi-Fi. Tota does not need any.
A glossary for doctors and clinicians
Tota ships with a Medical starter glossary of clinical terminology, and you can edit it or add your own. A glossary is a short list of terms the speech model is nudged towards while it transcribes, with sound-alike matching afterwards for anything it still mishears. A specialty glossary might add:
- apixaban
- metformin
- amoxicillin
- atrial fibrillation
- COPD
- HbA1c
- eGFR
- troponin
- laparoscopic
- bilateral
- ramipril
- colleague and consultant names
The live hint has room for roughly 70 to 90 words, so a specialty glossary works better than one list for all of medicine. Tota can switch glossaries by app, so the list used in the EHR can differ from the one used in email.
Tota ships with a built-in Medical starter glossary you can edit, and the glossaries guide covers set-up, the hint budget and per-app switching.
What Tota does not do
- It is not a medical device or an ambient scribe. Tota transcribes what you dictate. It does not listen to the consultation, summarise it or write the note for you, and it has no clinical certification.
- It is not HIPAA-certified and does not sign BAAs. A business associate agreement covers a vendor that handles protected health information. Tota does not handle it: patient text exists only on the Mac you dictated it on. Your organisation's device encryption, access and backup policies are what apply.
- It does not integrate with your EHR. It types into the EHR's text fields like a keyboard, which is usually what you want, but it cannot read or file into the record.
- Transcripts are files on your Mac. Nothing is synced or backed up by Tota. FileVault, your backup tool and your device policy are what protect them.
Pricing for a clinic
Tota is a one-time purchase of £19.99, with two seats per licence, after a 14-day free trial of every feature. There is no subscription, no per-minute transcription fee and no word limit, because local processing costs us nothing per word. Larger teams buy additional licences. Pricing is the same worldwide; see the pricing section for what the licence includes.
Frequently asked questions
Does patient information leave the Mac when I dictate?
No. Audio is transcribed on the Mac by OpenAI's Whisper model through WhisperKit, deleted once transcribed, and the text is typed into the app you are using. Tota sends no audio, text or screen content to any server. Its only network calls are a licence check, an update check and one-time model downloads.
Is Tota HIPAA compliant?
Tota makes no HIPAA claim and does not sign business associate agreements, because no Tota server handles patient information. Compliance of your overall setup depends on the Mac, the record system you dictate into and your organisation's policies. The security page enumerates every network call so your governance team can verify it.
Can it type into Epic, Cerner, EMIS or SystmOne?
If the field is a normal text field, including in a browser, yes. Tota is system-wide dictation and does not need an integration. Some record systems use custom controls that block all keyboard input, in which case Tota falls back to copying the text to the clipboard.
How accurate is it with drug names?
Whisper is good at ordinary English and weaker on rare drug and condition names. The Medical starter glossary and your own additions bias the model towards your vocabulary while it transcribes, and phonetic matching fixes sound-alike errors afterwards. Corrections you make are learned by the dictionary.
Does it work with no Wi-Fi?
Yes. After the one-time model download, Tota needs no connection at all. It works the same in a basement clinic, an ambulance bay or a home office.
Where are transcripts stored?
In a local history for three days, then archived as Markdown files in a folder you choose. Entries can be deleted at any time. Nothing is synced anywhere by Tota.
What does it cost?
£19.99 once, after a 14-day free trial, with two seats per licence. No subscription, no per-minute charge and no word limit.
Dictate the clinic letters on the Mac in the room
Download Tota, enable the Medical glossary, and clear the letter backlog without a governance form. Two seats, £19.99 once.
14-day free trial · macOS 14 or later · No account needed
