For Odoo 18.0 · Community & Enterprise · AminiTech HMS
Ambient AI Clinical Scribe
$149
$99
★ 30% LAUNCH OFFER
Your clinicians talk to patients â the AI writes the note. Capture the encounter audio, transcribe it with Whisper or Deepgram, and let Claude draft the SOAP, progress or discharge note before the patient leaves the room. Every draft passes through human-in-the-loop review, redaction rules and token quotas â DAX/Abridge-class scribing inside your HMS.
✅ Claude-powered drafting✅ Whisper/Deepgram transcription✅ Human-in-the-loop review✅ PHI redaction rules
⭐ 90 days free support
💻 Try the LIVE demo — login demo / demo
📚 Documentation
Speak your patients' language
Works in your language
Every screen, label and report is translated — switch a user's language in Odoo and the whole system follows.
EnglishFrançaisEspañolPortuguêsKiswahiliDeutschä¸æاÙعربÙØ©
8
dedicated models â recordings, scribe tasks, transcript segments, providers, templates, redaction rules, edit log and usage quotas.
9
encounter languages including English, Swahili, French, Arabic, Zulu, Yoruba, Hausa and Amharic â plus auto-detect.
3
LLM families supported for note drafting (Claude primary, GPT, Gemini) behind a pluggable provider registry.
📸 Live screens
Features
★ All features
Setup in 4 steps
★ Live demo ↗
HMS family
FAQ
Support & Contact
CLINICAL AI
What you get
Every feature below is live on the public demo — open hms.aminitechsolutions.com (login demo / demo) and walk it yourself.
01 · ENCOUNTER RECORDING
Capture the visit, not the keyboard
Each recording links the patient, clinician, encounter reference and start/end times, tracks duration automatically and supports 8 languages plus auto-detect â including Swahili, Zulu, Yoruba, Hausa and Amharic. Full mail-thread chatter and tracking on every recording.
02 · SCRIBE TASKS
Claude drafts the note in your format
Scribe tasks turn transcripts into SOAP, progress or discharge notes using configurable Scribe Templates. Claude is the primary LLM (Sonnet/Opus with per-model token pricing built in), with GPT and Gemini as alternatives â pick per provider, per task.
03 · HUMAN IN THE LOOP
Clinicians edit, the Edit Log remembers
No note is final until a clinician reviews it. Every change is captured in a dedicated Edit Log, so quality teams can see exactly what the AI wrote versus what was signed â the audit trail regulators and medical directors ask for.
04 · REDACTION RULES
PHI stays inside the hospital
Configurable Redaction Rules scrub sensitive identifiers before text ever leaves for an LLM. Ships with default redaction data and an admin-only Configuration menu so only your AI administrators can loosen the guardrails.
05 · USAGE & QUOTAS
Token budgets that stop bill shock
A Usage & Quota screen meters token consumption per period, with real per-1M-token pricing for Claude models baked in. Set quotas, watch spend, and let the scheduled cron keep processing queues moving in the background.
06 · PROVIDERS & REST API
Pluggable providers, open endpoints
Transcription (Whisper, Deepgram) and drafting providers are records, not hard-coded â swap keys and models from the Providers screen. A REST API lets your mobile capture app push encounter audio straight into the scribe pipeline.
Real screens — not mockups
See it running on the live demo
Every screen below is a real screenshot from the working system at hms.aminitechsolutions.com — log in with demo / demo and try it yourself.
Live screen recording — captured from the running demo.
Everything included
All 64 features — nothing held back
The complete capability list for Ambient AI Clinical Scribe, straight from the module. Every one of these is live on the demo at hms.aminitechsolutions.com.
Encounter Recording & Capture8 features
✓
Encounter recording records — Each clinician-patient encounter is captured as a recording record with patient, clinician, encounter reference, audio URL and MIME type.
✓
Auto-numbered recordings — Every recording gets a unique sequential reference (AI/ENC/...) from a dedicated ir.sequence at creation.
✓
8-language + auto-detect support — Recordings can be tagged English, Swahili, French, Arabic, Zulu, Yoruba, Hausa, Amharic or auto-detect for transcription.
✓
Automatic duration tracking — Encounter duration in seconds is computed and stored from start and end timestamps.
✓
Patient consent flag — A consent-recorded checkbox documents that the patient agreed to being recorded before AI processing.
✓
Retention date control — A retain-until date on each recording supports data-retention policy enforcement.
✓
Speaker-labelled transcript segments — Timed transcript segments carry speaker labels (clinician, patient, caregiver, other), confidence scores and a contains-PHI flag, editable inline.
✓
Chatter and activity tracking — Recordings inherit mail.thread and activities, with tracked changes to patient, clinician, timestamps, statuses and state.
AI Note Drafting (Scribe)7 features
✓
One-click SOAP drafting via Claude — A Draft SOAP action calls the Anthropic Messages API (claude-sonnet-4-5 default) with the transcript and an active SOAP template to generate the note.
✓
Configurable model parameters — Temperature, max tokens and system/user prompts come from the selected template and provider, sent verbatim in the API request.
✓
Graceful stub fallback — If no API key or active SOAP template is configured, a deterministic offline SOAP stub is generated so the workflow never blocks.
✓
Error-safe API handling — HTTP errors and exceptions are caught, logged on the task with the message, and marked failed without ever raising to the user.
✓
Real-time cost accounting — Input/output token usage from each API response is converted to USD cost using built-in per-model Anthropic pricing tables.
✓
Latency measurement — Every LLM call records wall-clock latency in seconds on its task for performance monitoring.
✓
Suggested billing codes field — A dedicated text field stores AI-suggested billing codes alongside the generated SOAP note.
Clinical Workflow & Sign-off5 features
✓
8-stage encounter state machine — Recordings move through Recording, Uploaded, Transcribing, Drafting, Draft Ready, Clinician Reviewing, Signed Off and Discarded with a statusbar.
✓
Stage-aware workflow buttons — Mark Uploaded, Start Transcription, Mark Draft Ready, Sign Off and Discard buttons appear only in valid states.
✓
Separate transcription and LLM statuses — Independent pending/in-progress/completed/failed statuses track speech-to-text and note-generation pipelines.
✓
Clinician sign-off gate — An explicit highlighted Sign Off action closes the human-in-the-loop review before the note is final.
✓
Human edit audit log — Each edit is logged with editor, timestamp, field changed, original vs new text, edit distance and accepted-AI flag for HITL benchmarking.
AI Task Ledger & Monitoring5 features
✓
Per-call scribe task records — Every LLM or transcription job creates a task record linking recording, provider and template with full prompt and response text.
✓
Task lifecycle states — Tasks track queued, running, succeeded, failed, cancelled and retrying statuses with start/finish timestamps and retry counts.
✓
Prompt/response/error inspection — Task form exposes the exact JSON prompt sent, the raw response and any error message in notebook tabs.
✓
Automatic failed-task retry cron — A 5-minute scheduled job requeues failed tasks (up to 3 retries, 50 per pass) into retrying state.
✓
Task cost and token analytics — Pivot and graph views break down cost, prompt tokens and completion tokens by provider and status.
Usage Quotas & Cost Control4 features
✓
Per-provider usage quotas — Quota records track requests made vs limit, tokens in/out and total USD cost per provider per period.
✓
Daily usage aggregation cron — A nightly job rolls up yesterday's tasks per provider into quota rows (requests, tokens, cost), creating or updating as needed.
✓
Quota threshold alerting cron — An hourly job posts a chatter alert once usage crosses the configurable threshold percentage (default 80%) and marks it sent.
✓
Cost pivot analysis — A pivot view slices total USD cost by provider and period for spend review.
AI Provider Registry7 features
✓
Multi-vendor provider catalog — 11 provider types: Anthropic Claude, OpenAI GPT, Azure OpenAI, Google Gemini, AWS Bedrock, Ollama, llama.cpp, Whisper local/API, Deepgram, Azure Speech.
✓
Capability routing — Each provider declares a capability (transcription, summarization, coding/billing, translation, dictation, multimodal) used to pick the right engine per job.
✓
PHI-safe (BAA/DPA) flag — Providers can be marked PHI-safe when a Business Associate or Data Processing Agreement exists, with a search filter to find them.
✓
Rate limits and defaults per provider — Requests-per-minute, temperature, max tokens, model name, endpoint and region are configurable per provider with unique-code enforcement.
✓
Key-never-stored design — Provider records carry only an API-key-configured boolean; the actual secret lives in system parameters, never on the record.
✓
8 pre-seeded providers — Ships Claude Sonnet 4.5, Claude Opus 4.1, GPT-4o, Azure OpenAI (HIPAA), Gemini 2.5 Pro, Bedrock Claude, Whisper API and Deepgram Nova-3 Medical.
✓
Claude-first provider selection — SOAP drafting prefers an active Anthropic summarization provider and falls back to any active summarization provider.
Prompt Template Library5 features
✓
Prompt template engine — Templates pair a system prompt with a user prompt using {{transcript}} and {{patient}} placeholders, plus per-template temperature and max tokens.
✓
8 note output formats — Templates target SOAP note, HPI only, discharge summary, operative note, referral letter, admission note, progress note or structured JSON.
✓
15 medical specialties — Templates scope to General, Paeds, OB/GYN, Surgery, ICU, Emergency, Cardio, Onco, Derm, Psych, Ortho, Uro, MCH, HIV/ART and TB.
✓
10 ready-made clinical templates — Ships SOAP, HPI, Discharge, Operative Note, Referral, Admission, Progress, Paeds SOAP, OB/GYN SOAP and strict-JSON extractor with anti-hallucination prompts.
✓
Per-template redaction requirement — A requires-PHI-redaction flag marks templates whose input must be de-identified first.
PHI Redaction & Privacy5 features
✓
Rule-based PHI redaction engine — Redaction rules combine regex, named-entity or list-lookup matching with custom replacement text and drag-to-order sequencing.
✓
9 PHI classes covered — Rules target patient name, DOB, MRN, phone, email, address, SSN, national ID or all PHI.
✓
Severity tiers and language scoping — Each rule is Required, Recommended or Optional and can be limited to any of 8 languages or auto.
✓
12 pre-seeded redaction rules — Ships US SSN, Kenya national ID, Nigeria NIN, Kenyan and E.164 phone, email, two DOB formats, MRN, street address, credit card and person-name NER rules.
✓
PHI-redacted flag on recordings — Recordings carry a redacted boolean so downstream users can see whether de-identification was applied.
REST API (JSON endpoints)6 features
✓
Start-recording endpoint — POST /hms/ai/recording/start creates a recording with patient, encounter reference, language and consent flag, returning its id and state.
✓
Stop-recording endpoint — POST /hms/ai/recording/<id>/stop closes the encounter, stamps the end time and returns the computed duration.
✓
Transcript-upload endpoint — POST /hms/ai/recording/<id>/transcript stores externally produced transcript text and advances the record to drafting.
✓
Draft-SOAP endpoint — POST /hms/ai/recording/<id>/draft_soap triggers the Claude drafting pipeline and returns the note, task id, state and any error.
✓
Sign-off endpoint — POST /hms/ai/recording/<id>/signoff finalizes the note via the API for integration workflows.
✓
Live usage dashboard endpoint — GET /hms/ai/usage/today returns JSON with today's quota rows plus a live per-provider aggregate of tasks, tokens and cost.
Views, Menus & Analytics5 features
✓
Ambient AI menu under HMS Clinical — A dedicated Ambient AI menu tree offers Recordings, AI Tasks (Scribe Tasks, Edit Log), Usage & Quota and an admin-only Configuration section.
✓
Recording kanban pipeline board — Recordings display as a kanban grouped by workflow state with patient, clinician and start time on each card.
✓
Recording pivot and graph analytics — Pivot (clinician x state, duration measure) and graph views analyze recording throughput.
✓
Rich recording form with tabs — The form organizes Transcript, Generated SOAP + billing codes, Segments, AI Tasks and Edits into notebook pages with chatter.
✓
Search filters and grouping — Recordings filter by Mine, Draft Ready, Signed Off and group by state, clinician, language; providers group by type or capability.
Security & Access Control4 features
✓
Two-tier role model — Ambient AI User and Ambient AI Admin groups (admin implies user) in a dedicated HMS Ambient AI category govern access via ir.model.access rules.
✓
Admin-only configuration menus — Providers, templates, redaction rules and settings menus are visible only to the Ambient AI Admin group.
✓
HMS phase-4 licence gating — All module menus are gated behind the HMS base consultation-ambient-AI phase group via the phase-4 gate function.
✓
Post-install admin bootstrap — A post-init hook automatically adds the superuser to both AI groups so the app works immediately after install.
Configuration & Setup3 features
✓
Settings panel for Anthropic key — An Ambient AI section in Settings stores the Anthropic API key as a password-masked system parameter.
✓
Deep HMS suite integration — Builds on aminitech_hms_base patients, smart-text and concept-dictionary modules and links recordings to hms.patient records.
✓
Live demo instance — The listing links a hosted live test environment at hms.aminitechsolutions.com for hands-on evaluation.
Setup in 4 steps
From install to first live workflow
1 · Install 2 · Configure providers
Install AminiTech HMS: Ambient AI Clinical Scribe. Required HMS apps are added at checkout automatically; default providers, templates and redaction rules load on install. Under Clinical > Ambient AI > Configuration, add your Anthropic API key (Claude Sonnet is the default model) and a transcription provider such as Whisper or Deepgram.
3 · Run a test encounter 4 · Go live
Create a Recording for a test patient, attach the encounter audio, run transcription, then launch a Scribe Task with a SOAP template and review the drafted note. Set usage quotas per clinician or department, confirm redaction rules match your privacy policy, grant the AI Admin group, and roll out to consulting rooms.
Save with a suite
Buy this app on its own, or get it in a suite
Buying more than a couple of apps? The Complete Suite gives you 170 modules for $3,999 in one purchase — usually less than the apps bought separately, and everything is installed and updated together.
BEST FIT FOR THIS APP
Complete Suite
$3,999
170 modules — the entire HMS catalogue
✓ includes this app
View the Complete Suite ↗What's in your download
AI Clinical Suite installs with everything it needs
This app builds on the apps below. Odoo adds them to your cart automatically and delivers everything in one download — you are only charged for apps you don't already own.
Adds these apps to your cart automatically
✓ Plus 30 more modules included free — the code that powers every feature, delivered in the same download.
Fitting out a whole facility? The Complete Suite includes this app and its whole family in one purchase.
Every included module
Every feature of every module in this app
This one purchase includes 3 modules with 156 features in total — each module's full capability tour below.
{}36 features included in this app
Ambient AI Documentation9 features
✓
Ambient AI scribe sessions — Dedicated hms.scribe.session records capture a full recording-to-note workflow, ordered newest-first with auto-generated references.
✓
Consultation audio capture — Attach the recorded doctor-patient conversation as a binary audio file (stored as an attachment) with its filename on each session.
✓
Whisper speech-to-text transcription — One-click Transcribe posts the audio to a Whisper-compatible endpoint (OpenAI whisper-1 by default) and stores the returned transcript on the session.
✓
LLM SOAP note synthesis — One-click Synthesise sends the transcript to an LLM (Anthropic Claude by default) and parses a strict JSON reply into structured note fields.
✓
Structured SOAP fields — Separate Subjective, Objective, Assessment and Plan text fields hold the AI-drafted note for clinician review before anything touches the chart.
✓
ICD-10 code suggestions — The LLM returns comma-separated suggested ICD-10 codes, stored on the session alongside the SOAP draft.
✓
Apply-to-consultation autofill — A third button copies the reviewed SOAP sections into the consultation's complaint, examination, assessment and plan fields automatically.
✓
Defensive field mapping — Apply logic probes multiple candidate consultation field names (complaint/chief_complaint, diagnosis/impression, etc.) so it fits HMS variants.
✓
Graceful stub mode without API keys — With no keys configured, transcribe and synthesise return clearly-labelled stub content so the module installs cleanly and the workflow is demoable.
Workflow & State Machine5 features
✓
Seven-stage session lifecycle — Sessions move draft to transcribing, transcribed, synthesising, ready and applied, with a cancelled state, shown on a form statusbar.
✓
Numbered guided workflow buttons — Header buttons labelled 1) Transcribe, 2) Synthesise SOAP, 3) Apply to Consultation appear only in the valid state, guiding staff step by step.
✓
Cancel action — Any unfinished session can be cancelled in one click without touching the consultation record.
✓
Failure recovery with audit trail — Transcription or synthesis errors are caught, logged to the chatter as a message, and the session rolls back to its previous state for retry.
✓
Transcript-first guardrail — Synthesis refuses to run until a transcript exists, and Apply refuses until a consultation is linked, preventing empty notes.
Consultation Integration4 features
✓
AI Scribe smart button on consultations — A microphone stat button on the consultation form shows the session count and opens the scribe workflow in one click.
✓
One-click session creation from the chart — Opening the scribe from a consultation auto-creates a draft session pre-linked to that consultation.
✓
Linked patient and doctor context — Patient and doctor are related from the consultation, stored and tracked on the session, and shown read-only on the form.
✓
Session history per consultation — A one2many keeps every scribe session attached to its consultation with a computed session counter.
REST API7 features
✓
Create-session endpoint — POST /scribe/session creates a draft session (optionally linked to a consultation) and returns its id, reference and state as JSON.
✓
Audio upload endpoint — POST /scribe/session/<id>/audio accepts multipart audio from any recorder client and stores it on the session, returning byte count.
✓
Transcribe endpoint — POST /scribe/session/<id>/transcribe runs Whisper server-side and returns state, transcript length and a preview.
✓
Synthesise endpoint — POST /scribe/session/<id>/synthesise runs the LLM and returns SOAP assessment and plan previews as JSON.
✓
Apply endpoint — POST /scribe/session/<id>/apply pushes the note into the consultation via the API, for kiosk or mobile front-ends.
✓
Session status endpoint — GET /scribe/session/<id> returns JSON state including has_audio, transcript length and soap_ready flags for polling clients.
✓
Authenticated JSON API with error handling — All endpoints require an authenticated user session and return clean JSON errors (not_found, no_audio) with proper HTTP status codes.
Views & Usability5 features
✓
Tabbed session form — Notebook pages separate Audio, Transcript and SOAP Draft so the raw recording, verbatim text and structured note each get a clean workspace.
✓
Session list view — A list view shows reference, consultation, patient, doctor, state and creation date for scanning all scribe activity.
✓
AI Scribe Sessions window action — A ready-made act_window (list + form) exposes all sessions for browsing and administration.
✓
Duration and cost tracking — Each session records audio duration in seconds and an estimated provider cost in cents.
✓
Chatter and activities on every session — Sessions inherit mail.thread and mail.activity.mixin: full message log, field tracking on state/patient/doctor, and follow-up activities.
Configuration & Security6 features
✓
System-parameter configuration — Whisper endpoint/model/key and LLM endpoint/model/key are all plain system parameters, so providers can be switched without code changes.
✓
Provider-agnostic endpoints — Both the transcription and LLM endpoints are overridable, so self-hosted Whisper or alternate LLM gateways can be plugged in.
✓
Seeded default parameters — noupdate data pre-seeds OpenAI Whisper and Anthropic Messages endpoints with sensible default models on install.
✓
Automatic session numbering — An ir.sequence issues references like SCB/2026/00001 per session, seeded by the module.
✓
Access rights for internal users — ir.model.access grants internal users full CRUD on scribe sessions out of the box.
✓
Multi-company aware — Each session carries a company_id defaulting to the active company.
{}56 features included in this app
Conversational EHR Interface6 features
✓
Talk to the EHR in natural language — A full-screen chat client where doctors type or dictate plain-English questions and commands against live patient, lab, Rx and triage data.
✓
Three-pane chat workspace — Dedicated OWL screen with a conversation sidebar, central message stream and a right-hand patient context panel.
✓
ChatGPT-style message bubbles — User and assistant turns render as distinct role-labelled bubbles in a scrolling conversation pane.
✓
Optimistic send with auto-scroll — Your message appears instantly and the pane auto-scrolls to the newest reply while the server responds.
✓
Empty-state example prompt — A fresh chat shows a worked example ('Show me my patients with HbA1c > 9 not on insulin') to teach query style.
✓
Enter-to-send composer — Enter sends and Shift+Enter inserts a newline in the multi-line input box, which auto-focuses on load.
Natural-Language Intent Engine5 features
✓
Rule-based intent parser — A regex and keyword engine classifies roughly 30 common clinical query patterns without needing an LLM key.
✓
LLM fallback via ambient scribe — Unrecognised queries are dispatched to hms.ai.scribe.task, degrading gracefully to a canned hint when no AI provider is configured.
✓
Parsed-intent cache — hms.chat.intent stores text hashes, raw text, resolved intent, params and a hit counter to reuse prior parses.
✓
Multi-turn refinement — Follow-ups like 'add zinc to that order' or 'and metformin too' attach to the previous assistant tool call for continued editing.
✓
Confidence scoring — Each parsed reply carries a confidence value alongside its intent and parameters.
Clinical Query Intents11 features
✓
Triage queue count — 'How many in triage' returns today's triage-queue headcount as a metric card.
✓
HbA1c threshold with medication exclusion — Parses 'HbA1c > 9 not on insulin', filtering lab results above a threshold while excluding patients on a named drug.
✓
Generic lab-value filter — Any 'test <op> value' phrase filters lab results numerically across >, >=, < and <= comparisons.
✓
Latest patient vitals lookup — 'Patient 42 vitals' returns that patient's most recent BP and temperature as a vitals card.
✓
Open patient chart by name — 'Open Jane's chart' searches patients by name fragment and offers one-click chart openers.
✓
Draft prescription from a sentence — 'Prescribe amoxicillin 500 mg TDS patient 42' parses drug, dose, unit, frequency and patient into an editable Rx draft.
✓
Today's no-shows — Lists cancelled consultations for the day as a patient list card.
✓
Lab criticals in last 24h — Surfaces lab results flagged critical-high, critical-low or panic in a lab-result list card.
✓
Patients on a given medication — 'Patients on warfarin' matches prescription lines by generic name or product to list affected patients.
✓
Patient search by name fragment — 'Find patients named Smith' returns matching charts for quick opening.
✓
Roadmap placeholder intents — Overdue follow-ups, ICU census, OR schedule, discharge-ready and pending-refills are recognised and answer with wire-up placeholders.
Tool-Call Result Cards6 features
✓
Patient list card — Renders query results as a clickable list of patients that deep-link to their charts.
✓
Lab result list card — Displays matching lab results with test, value, date and patient for one-click drilldown.
✓
Vitals card — Shows a patient's latest BP, temperature and capture time in a compact card.
✓
Metric card — Presents single-number answers such as triage counts as a labelled value tile.
✓
Editable Rx draft card — Prescription drafts render with per-line drug, dose, unit, frequency, duration and route fields you can edit before approving.
✓
Add/remove Rx lines inline — The Rx draft card lets clinicians add or delete medication lines before committing the order.
Explicit-Approval Mutations5 features
✓
Approve-to-execute gate — Data-changing tool calls (like Rx drafts) only run after an explicit approve click, never automatically.
✓
One-click reject — Rejecting a tool call marks the message rejected and performs no mutation.
✓
Prescription commit to pharmacy — Approving an Rx draft creates a real hms.prescription with lines, mapping frequency/route/unit shorthand to the pharmacy's own selection values.
✓
Auto-link consultation and product — On commit it finds or creates the required consultation and matching product record so the prescription is valid end-to-end.
✓
Approval audit stamp — Messages record approved/rejected flags and an executed_at timestamp for each tool call decision.
Conversation Management7 features
✓
Persistent conversations — Every chat is stored as hms.chat.conversation with its full ordered message history.
✓
Auto-titled from first message — A new conversation is renamed from the opening user message, truncated with an ellipsis.
✓
Conversation sidebar — Lists your open chats with pin markers, last-activity time and message counts, capped at 80.
✓
Pin conversations — Conversations can be pinned to keep them prominent in the sidebar.
✓
Rename and archive — Chats can be renamed or archived (open/archived state) to keep the list tidy.
✓
Share with colleagues — A conversation can be shared with other users via a many-to-many share relation.
✓
Message-count rollup — Each conversation computes its live message total for the sidebar.
Patient & Visit Context4 features
✓
Sticky patient context — A conversation carries a current patient and visit so follow-up questions stay scoped without re-stating the name.
✓
Context side panel — The right panel shows the active patient's name, DOB and sex with an 'Open chart' button, or a prompt to pick one.
✓
Set-context endpoint — Clicking a patient in a result card sets conversation context and can open their chart in the main view.
✓
Citation drilldowns — Assistant replies list clickable source records (labs, patients, vitals) that open the underlying form in a dialog.
Voice, Streaming & Notes4 features
✓
Voice input (speech-to-text) — A mic button uses the browser SpeechRecognition API to dictate queries into the composer, with graceful fallback if unsupported.
✓
Server-sent streaming replies — An SSE endpoint streams the assistant response word-by-word for a live typing effect.
✓
Save conversation as chart note — 'Save as Note' renders the transcript to an HTML document and attaches it to the patient's chart as an ir.attachment.
✓
Token usage gauge — The header shows a live tokens-used-versus-limit bar sourced from the shared AI usage quota.
Quick Prompts & Configuration2 features
✓
One-click quick prompts — Chip buttons fire preset queries such as no-shows, lab criticals, ICU census and pending refills.
✓
Configurable prompt catalog — Quick prompts are seeded in a system parameter and can be edited to any JSON list without code changes.
API, Security & Integration6 features
✓
JSON chat API — Endpoints for send, history, quick-prompts, execute-tool-call, save-as-note and set-context back the whole UI over authenticated JSON routes.
✓
User-scoped access rules — Conversation, message and intent models are readable and writable by all internal users via dedicated access rules.
✓
HMS clinical menu integration — Adds a 'Talk to EHR' entry under the HMS Base clinical menu as a client action.
✓
Deep HMS suite integration — Reads across consultation, laboratory, pharmacy, concept dictionary, CDS, ambient-AI and smart-text modules for its answers.
✓
Phase-gate self-check — Ships a phase-4 gate data file asserting its clean install against hms_base and chat dependencies.
✓
Swahili translation — Includes an sw.po catalog so the interface strings can render in Swahili.
One patient record, one family
The AminiTech HMS family
60+ hospital apps that share one patient record and install together. Required apps are added to your cart automatically at checkout — you only ever download what you have paid for. Or get everything at suite pricing with the Complete Suite.
INCLUDED WITH THIS APP
Ambient AI Scribe (DAX / Abridge equivalent)
Record a consultation conversation, get a structured SOAP note auto-filled into the consultation form. Uses Wh
INCLUDED WITH THIS APP
Chat (Conversation-Driven EHR)
Talk to the EHR like Claude. Natural language queries, tool-call cards, voice, streaming.
INCLUDED WITH THIS APP
Voice OR (Hands-Free Operating Room)
Voice-driven, hands-free OR cockpit. Wake-word "Bridge" + TTS responses for sterile surgeons.
INCLUDED WITH THIS APP
- AI-First Clinical (Doctor Approves)
AI drafts the entire encounter (SOAP, orders, Rx, follow-up, billing) - doctor reviews & approves.
FAQ
Common questions
What other apps does this module require?
It depends on AminiTech HMS Base, Smart Text and Concept Dictionary. You don't need to buy them separately by hand â required apps are added to your cart automatically at checkout on the Odoo Apps store.
Does it work on Odoo Community and Enterprise?
Yes â both Community and Enterprise editions of Odoo 18 are supported. It cannot be installed on Odoo Online (SaaS), since custom modules require an on-premise or Odoo.sh-style deployment.
Do I need my own AI API keys?
Yes. You bring your own keys for the LLM (Claude is the primary and default; GPT and Gemini are also supported) and for transcription (Whisper or Deepgram). Built-in quotas and per-token pricing help you track exactly what each key spends.
Is patient data safe when sent to the AI?
Redaction rules scrub configurable identifiers before text is sent to any provider, access is gated by dedicated security groups, and every AI draft plus every human edit is logged. The clinician always signs the final note.
What support do you offer?
Every purchase includes a 90-day bug-fix guarantee. For help, customization or integration questions, email reach@aminitechsolutions.com â we respond fast and support real deployments.
Support & Contact
Talk to us today
🏆
90 days of free support included with your purchase
Real people, business hours, typical first reply under 4 hours — including installation help and configuration for your hospital.
⭐ Included — no extra cost
📞
Phone / WhatsApp
+254 746 883 809
💻
Live demo
hms.aminitechsolutions.com
📚
Documentation
aminitechsolutions.com/docs
Odoo Proprietary License v1.0 This software and associated files (the "Software") may only be used (executed, modified, executed after modifications) if you have purchased a valid license from the authors, typically via Odoo Apps, or if you have received a written agreement from the authors of the Software (see the COPYRIGHT file). You may develop Odoo modules that use the Software as a library (typically by depending on it, importing it and using its resources), but without copying any source code or material from the Software. You may distribute those modules under the license of your choice, provided that this license is compatible with the terms of the Odoo Proprietary License (For example: LGPL, MIT, or proprietary licenses similar to this one). It is forbidden to publish, distribute, sublicense, or sell copies of the Software or modified copies of the Software. The above copyright notice and this permission notice must be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Please log in to comment on this module