For insurance-based therapy practices

$7,200/year/clinician in new revenue.

At $50/month. Your first 60 days are free.

Lucid is a Mac desktop app that runs passively during your therapy sessions and collects brief voice check-ins from patients 2-3 times per week. Each touchpoint automatically generates a CPT-billable mental health assessment. Your clinical workflow doesn't change.

Start My Free Trial See the Research

14,898-patient validation Published, Annals of Family Medicine HIPAA on-device, no BAA 72% payer reimbursement

Click to watch overview

How It Works

Two billing events per patient, per week.

One during the session you're already having. One each time they check in between sessions. Both generated automatically.

Download Lucid on your Mac

Five-minute setup. No EHR integration required, no new hardware, no IT department. Works alongside whatever you're already using.

Live today

Open Lucid before each session

It listens passively. Your session runs exactly as it always has. By the time the session ends, Lucid has generated a full voice biomarker assessment — PHQ-9 correlated, CPT-documented.

Zero overhead per session

Patients check in between sessions

2-3 times per week, each check-in under 2 minutes. They speak into their browser — no app to download, no account to create. Each check-in generates another billable assessment.

Another CPT code each time

Patient Workflow

What your patients experience

Six simple steps. No behavior change required. Lucid works quietly in the background while your patients go about their day.

Download Lucid

Your patient downloads Lucid to their Mac. One installer, takes under a minute. No account creation, no email, no personal information collected.

Quick Setup

A 2-minute onboarding: grant microphone permission, enroll their voice with a short reading passage. That's it — Lucid is ready.

Just Talk Normally

Lucid listens passively throughout the day — calls, meetings, conversations. It analyzes voice patterns in real time and updates the dashboard with a composite Voice Score, clinical indicators (PHQ-9, GAD-7), and five voice dimensions.

Lucid dashboard showing Voice Score of 66, clinical indicators, and voice dimensions

Track Progress Over Time

Patients can view their trends — daily, weekly, monthly, or yearly. The Trends view shows their average score, best day, total readings, current streak, and clinical indicator trajectories.

Lucid trends view showing 7-day Voice Score trend and clinical indicators

Share with Their Therapist

When ready, the patient enters your pairing code in Settings. From that point forward, Lucid automatically sends encrypted pre-session reports before each appointment. The patient can revoke access at any time.

Complete Privacy

All voice processing happens entirely on the patient's device. No audio is ever transmitted, stored in the cloud, or accessible to anyone — including you. Only derived scores and patterns are shared, and only when the patient explicitly opts in.

Therapist Workflow

What you see as the provider

Objective voice data delivered before every session. Review in 60 seconds, bill two units, and walk into session with data-informed talking points.

Patient Connects

Give your patient a 5-character pairing code. They enter it in their Lucid app. You receive a confirmation that they've linked — no software to install on your end, no portal to manage.

Automatic Pre-Session Brief

Two hours before each scheduled appointment, you receive a Session Ready Brief via secure email. It's a single-page summary: Voice Score with 14-day trend, PHQ-9 and GAD-7 gauges, flagged patterns, and suggested conversation starters derived from the data.

Session Ready Brief showing Voice Score, clinical snapshot, flagged patterns, and session talking points

60-Second Review

Scan the brief before your patient walks in. The layout is designed to read in under a minute: the Voice Score and trend tell you direction, the clinical gauges show severity, and the talking points give you an evidence-based way to open the conversation.

"Your stress score improved significantly since last session — that's real progress. But your data shows it's still much higher on weekdays than weekends. What's different about your weekdays right now?"

Data-Informed Sessions

Instead of relying solely on self-report, you have objective between-session data. Patients often under-report improvements or miss patterns they can't see. The data gives you a more complete picture and helps surface what matters most.

Clinical Report for Documentation

Access the full 3-page clinical report: Voice Score summary with confidence intervals, emotion profile analysis, linguistic markers table, notable patterns detected, and a complete billing documentation page with CPT code, outputs reviewed, and signature lines.

Clinical report showing Voice Score summary, clinical indicators, and voice dimensions

Bill CPT 96127 x 2

Each patient session generates two billable units of 96127 (Brief emotional/behavioral assessment). The documentation is pre-filled with assessment tool description, specific outputs reviewed, and clinical action checkboxes. Add your interpretation, sign, and submit.

Billing Codes

The exact CPT codes Lucid generates.

Your billing team submits these through your existing workflow. Lucid generates the documentation. You don't need to change your billing process, your EHR, or your payer contracts.

Code	Description	Reimbursement	When Generated
96127	Brief behavioral or emotional assessment — standardized instrument, scored report	$4.97 (Medicare)	Every session
99484	Behavioral health integration (BHI) care management — 20+ min/month	$55 (Medicare)	Monthly, if criteria met
99492	Initial psychiatric collaborative care management — first 70 min/month	$155–235	First month of enrollment
99493	Subsequent psychiatric collaborative care management — 60 min/month	$105–165	Ongoing monthly
99494	Add-on: each additional 30 min of BHI care management	$50–80	When applicable

The Software

What Lucid generates for each patient.

After each session and each between-session check-in, Lucid produces a scored assessment with Depression Risk (PHQ-9 correlated), Anxiety Risk (GAD-7 correlated), and voice dimensions. That assessment is what generates the CPT documentation.

Per-session voice analysis

Lucid session dashboard — voice score 66/Steady, Depression Risk 9.3 Minimal, Anxiety Risk 8.3 Mild, Voice Dimensions: Energy 60, Stress 75, Focus 67.1, Mood 66, Recovery 60.7

Patient trends over time

Lucid trends view — 7-day average 68, 126 readings, Depression PHQ-9 10.1 Minimal, Anxiety GAD-7 8.5 Mild, with line chart

Clinical Validation

Peer-reviewed. Double-blind. 14,898 patients.

The underlying model was evaluated in a cross-sectional study across 14,898 English-speaking adults in the United States and Canada. Participants provided free-form speech of 25+ seconds and completed a self-reported PHQ-9.

At a cut-off score of 10 (moderate to severe depression), the voice biomarker model achieved sensitivity of 71.3% (95% CI: 69.0-73.5) and specificity of 72.5% (95% CI: 71.5-73.5) — comparable to clinician-administered screening instruments.

The study was conducted by Kintsugi Mindful Wellness. Data collected February 2021 through July 2022. Published January 2025.

14,898

Patients validated

71.3%

Sensitivity

72.5%

Specificity

244M

Model parameters

Mazur A, Costantino H, Tom P, Wilson MP, Thompson RG. Evaluation of an AI-Based Voice Biomarker Tool to Detect Signals Consistent With Moderate to Severe Depression. Ann Fam Med. 2025 Jan 27;23(1):60-65.

doi: 10.1370/afm.240091 · PMID: 39806690 · PMC11772039

PubMed entry for the Annals of Family Medicine study validating Kintsugi Voice v1 across 14,898 patients

The Math

What this looks like at different practice sizes.

These numbers assume standard payer reimbursement at current rates. Your actual revenue depends on your payer mix and patient panel.

Solo Practice

$649

new revenue per month

10 BHI patients, 1 clinician

Annual new revenue: $7,788

Lucid cost: $50/month

Payback: Day 1

13x ROI

Group Practice

$6,494

new revenue per month

10 clinicians, 100 BHI patients

Annual new revenue: $77,928

Lucid cost: $390/month

Payback: Day 1

17x ROI

Multi-Site Practice

$16,235

new revenue per month

25 clinicians, 250 BHI patients

Annual new revenue: $194,820

Lucid cost: $975/month

Payback: Day 1

13x ROI

Pricing

Straightforward pricing. No surprises.

Solo Practice

$50

per seat, per month

Unlimited in-session monitoring
Unlimited between-session check-ins
CPT documentation for all touchpoints
HIPAA-compliant, on-device processing
Cancel anytime, no contracts

Start 60-Day Free Trial

Group Practice

$39

per seat, per month

Everything in Solo
Practice-level dashboard
Bulk CPT reporting by clinician
Priority onboarding support
Volume pricing at 10+ seats

Start 60-Day Free Trial

60-day free trial. No credit card required. If you don't bill at least $50 in new CPT revenue in the first 60 days, we'll extend your trial.

Questions

Things practices ask before signing up.

No. The between-session check-in is a web page — patients access it through their browser on any device. No app, no account, no install. You send them a link; they click it and speak for 60-90 seconds.

Yes, by architecture. Lucid processes everything on your Mac — no audio is transmitted to any server, no data leaves the device. There's no cloud component, which means there's no BAA required and no third-party data sharing to disclose to patients.

72% of major commercial payers and Medicare currently reimburse 96127 and the 994xx behavioral health integration codes. This includes most BlueCross BluShield plans, Aetna, Cigna, UnitedHealth, and Medicare Part B. The specific reimbursement rates vary by state and payer. Lucid doesn't guarantee any payer's reimbursement decision — we provide the compliant documentation; billing is handled through your existing process.

Claim denials happen in every billing workflow. Lucid generates documentation that meets the documentation requirements for each code — the rest is between you and your payer. If you're not currently billing these codes at all, even a 50% approval rate on new CPT revenue is net positive versus $0 today.

You open Lucid before the session starts. That's it. The analysis happens in the background. You don't need to pause the session, enter scores, or do anything differently. The average additional time per session is under 2 minutes — typically the time it takes to close the software after the patient leaves.

Lucid doesn't integrate directly with EHRs. It generates CPT documentation that your billing team enters into whatever system you use. This keeps the integration simple and means Lucid works with any EHR — or no EHR.

Nothing is stored after analysis. Lucid processes audio in real time on your Mac and discards the raw audio immediately. The only thing that persists is the scored assessment output — the numbers, not the voice. No recordings are kept, transmitted, or stored anywhere.