Loading SaaS profile...
Kintsugi
Kintsugi is an AI-powered healthcare platform that analyzes voice biomarkers to screen for clinical depression and anxiety from short clips of natural speech. Built for healthcare providers, payors, telehealth vendors, and call centers, the platform detects early signs of mental health conditions regardless of native language or spoken context. By embedding directly into patient intake and call center workflows via API, Kintsugi supplies objective behavioral health data. This enables clinical teams to escalate care and route patients to appropriate interventions.
What it does
Analyzes 20-second clips of free-form speech to generate real-time depression and anxiety risk scores. Integrates via REST API with clinical EHR systems, telehealth portals, and call center software. Provides non-contextual vocal analysis that measures acoustic properties rather than spoken language or keywords. Delivers objective behavioral health metrics to triage high-risk patients for targeted clinical intervention.
Why it works
Language-agnostic acoustic analysis detects vocal micro-patterns regardless of primary dialect or topic. API-first design seamlessly embeds mental health screening directly into existing clinical software workflows. Objective voice biomarkers eliminate manual screening questionnaire bias and reduce patient drop-off. Proprietary dataset gathered across global users ensures diverse ML model validation and accuracy.
Category: AI & Automation
Team Size: 11-25
Visit Website
As of 2026
AI & Automation
Revenue: $1M-$10M ARR
AI voice biomarker platform identifying signs of depression and anxiety in real time.
Overview
Kintsugi is an AI-powered healthcare platform that analyzes voice biomarkers to screen for clinical depression and anxiety from short clips of natural speech. Built for healthcare providers, payors, telehealth vendors, and call centers, the platform detects early signs of mental health conditions regardless of native language or spoken context. By embedding directly into patient intake and call center workflows via API, Kintsugi supplies objective behavioral health data. This enables clinical teams to escalate care and route patients to appropriate interventions.
Funding status: Funded
Revenue source: Subscriptions
Customer type: B2B
Pricing: Enterprise
Tech stack: Python, PyTorch, TensorFlow, React, FastAPI, AWS, Docker, Signal Processing
Platform: Web • iOS • Android
Integrations: Pegasystems, REST API, Webhooks
Founder story
Kintsugi was founded in 2019 in Berkeley, California, by Grace Chang and Rima Seiilova-Olson. Both founders experienced challenges accessing timely mental healthcare and recognized that objective screening tools were largely absent from primary care. As engineers, they treated behavioral screening as an infrastructure problem and developed AI algorithms capable of detecting signs of clinical depression and anxiety from subtle vocal nuances.
What it does
- Analyzes 20-second clips of free-form speech to generate real-time depression and anxiety risk scores
- Integrates via REST API with clinical EHR systems, telehealth portals, and call center software
- Provides non-contextual vocal analysis that measures acoustic properties rather than spoken language or keywords
- Delivers objective behavioral health metrics to triage high-risk patients for targeted clinical intervention.
Who it's for
- Healthcare Insurance Payors & Health Plans
- Telehealth & Remote Patient Monitoring Providers
- Healthcare Call Centers & Nurse Triage Lines
- Health Systems & Behavioral Health Clinics
Why it works
- Language-agnostic acoustic analysis detects vocal micro-patterns regardless of primary dialect or topic
- API-first design seamlessly embeds mental health screening directly into existing clinical software workflows
- Objective voice biomarkers eliminate manual screening questionnaire bias and reduce patient drop-off
- Proprietary dataset gathered across global users ensures diverse ML model validation and accuracy.
Growth strategies
- Partnering with major enterprise healthcare platforms like Pegasystems to embed voice screening in payer call centers.
- Securing clinical grants from organizations like the National Science Foundation to validate medical efficacy.
- Using a consumer self- care journaling app to collect diverse, global speech datasets for AI model training.
- Co- selling with health plan payors seeking to identify undiagnosed behavioral health conditions early.
Alternatives
Comparison overview
- Kintsugi analyzes non-contextual voice biomarkers rather than transcribing spoken speech content, whereas natural language tools depend on specific text keywords or responses.
- Kintsugi delivers real-time screening scores via an API embedded in call center platforms like Pega, unlike standalone diagnostic apps requiring separate provider logins.