Mental health AI scribe
A scribe built for therapy, not the average medical visit.
General-purpose scribes transcribe any clinician. Citt.ai is tuned for behavioral health, and the note is only the entry point to a clinical intelligence platform.
Citt.ai vs general-purpose AI scribes
Tools like Heidi and Freed are built to document any medical encounter. That breadth is a strength in a busy clinic and a limitation in a therapy room. Here is where the two differ.
| Dimension | Citt.ai | General-purpose scribe |
|---|---|---|
| Built for | Mental health and therapy specifically | Any medical specialty, broadly |
| Note types | DAP, behavioral-health SOAP, biopsychosocial intake, EMDR, couples and family | Standard medical SOAP and a few generic formats |
| Risk and safety language | Captured precisely, treated as first-class, escalation paths in the workflow | Summarized like any other content, not specifically tuned |
| Modality awareness | Reflects how the clinician actually works | Tends to default to medical framing |
| After the note | Feeds a compiled patient picture, timeline, and decision support | Note is filed; usually nothing downstream |
| Continuity | Session connects to the long therapeutic arc | Each encounter is treated in isolation |
| Consent and provenance | Built in; summaries trace back to their sources | Varies; often a generic consent checkbox |
| Positioning | Clinician-supervised decision support and documentation | Documentation tool |
The scribe is the entry point, not the product
Transcription is becoming a commodity. The lasting value is what the system builds once the note exists, all under your oversight.
Compiled patient picture
Approved notes, assessment trends, check-ins, goals, and risk signals are pulled into one therapist-facing summary, with uncertainty preserved rather than hidden.
Patient timeline
Sessions, assessments, check-ins, and milestones become a chronological view, so the long arc is visible instead of buried across separate notes.
Citt Evidence
Guideline-grounded clinical decision support, so a clinical question can be answered against evidence and the patient's own context in the same workflow.
Read the full breakdown in why a mental health AI scribe beats a general-purpose one.
The same moment, documented two ways
A client says, near the end of a session: “I’ve been thinking it would be easier if I just wasn’t around, but I’d never actually do anything.” How that line is captured is where a behavioral-health scribe earns its place.
General-purpose scribe
S: Patient reports low mood this week.
O: Appears tired, engaged in session.
A: Depressive symptoms, stable.
P: Continue weekly sessions.
The most important clinical moment in the hour has been summarized away.
Citt.ai
Risk: Passive suicidal ideation expressed (“easier if I wasn’t around”); denies intent or plan.
Action: Risk assessment documented; safety plan reviewed.
Flagged for therapist review before note approval.
The clinical shape is preserved, surfaced for your review, and connected to the patient’s ongoing picture.
Illustrative example. Citt.ai does not make clinical decisions; the therapist reviews and approves every note.
Risk and safety content is captured precisely and stays inside the care workflow. You review and approve every note. Citt.ai is clinician-supervised decision support, not an autonomous therapist. See our trust and safety standards.
See it in your own practice
Explore the platform built around therapist-led care, with scribing as the entry point to clinical intelligence.