An open-source AI medical scribe that records patient encounters and generates structured clinical notes automatically.
OpenScribe is an open-source AI medical scribe that automates clinical documentation. It records patient encounters, transcribes audio using models like Whisper, and generates structured draft clinical notes using LLMs such as Claude, helping clinicians save time on administrative tasks.
Clinicians, healthcare providers, and medical practices looking for an open-source, privacy-focused tool to automate the creation of clinical notes from patient encounters.
Developers choose OpenScribe for its strong emphasis on data privacy, local-first architecture, and flexibility—offering multiple deployment modes (web, desktop, fully local) without vendor lock-in, all under an MIT license.
OpenScribe is an open-source AI scribe that records patient encounters and generates structured clinical notes automatically. You keep full control over data, workflows, and patient privacy with no vendor lock-in.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Encrypts data locally with AES-GCM and has no analytics or telemetry, ensuring sensitive patient information stays secure on the user's device, as highlighted in the privacy section.
Supports mixed web, fully local desktop, and cloud-based workflows, allowing users to balance privacy, cost, and performance, with detailed setup guides for each mode.
Components like LLM providers and transcription services are swappable, enabling customization and integration with different AI models, as described in the architecture.
Includes detailed setup guides, architecture diagrams, and contribution guidelines, making it easier for developers to adopt and contribute to the project.
The README explicitly states it's not yet HIPAA compliant, requiring health systems to perform additional audits and steps before production use with PHI.
It's a standalone tool with no direct connections to EHR systems, necessitating manual data entry or custom development for integration, as noted in limitations.
Requires configuration of environment variables, installation of multiple dependencies like Node.js and pnpm, and understanding of runtime modes, which can be daunting for non-technical users.