Who
You
This is a personal learning tool for anyone.
What
Anything
Learn from existing curriculum sources or topics of your own.
Where
Anywhere
A mobile app and ecosystem designed for learning on the go.
When
Anytime
Driving, chores, going for a walk.
How
Voice
AI-driven, fully interactive sessions that adapt to you.
See It In Action
These screenshots are from fully functional iOS builds currently in development. This is working software, not mockups or design concepts. The visual design is actively evolving.
Why UnaMentis?
UnaMentis delivers structured, high-quality curriculum through natural voice conversation. At any point, you can stop to ask questions, explore tangents, or request examples, then seamlessly return to exactly where you left off. Built for extended learning sessions with a provider-agnostic architecture that supports cloud services or fully self-hosted deployments for complete privacy.
Voice-Native Design
Built for voice from day one. Natural interruption handling, turn-taking logic, and voice activity detection create fluid conversations.
Provider Agnostic
Swap STT, TTS, and LLM providers without code changes. 9 STT providers, 5 TTS providers, 5 LLM providers including cloud, on-device, and self-hosted options.
Curriculum System
UMCF (UnaMentis Curriculum Format) is a purpose-built specification for conversational AI tutoring with voice-optimized content.
Low Latency
Natural conversation flow through careful architecture, prefetching, and intelligent routing.
Self-Hosted Ready
Run your own STT, TTS, and LLM servers for privacy-first deployments. Full support for Ollama, llama.cpp, Piper, vLLM, and more.
Works Anywhere
Graceful degradation means the app works on any device, even without API keys. Built-in fallbacks to Apple Speech, Apple TTS, and on-device LLM.
Full Observability
Real-time telemetry, cost tracking per provider, latency metrics, and thermal monitoring for production deployments.
UMCF Curriculum Format
UMCF (UnaMentis Curriculum Format) is a JSON-based curriculum specification designed specifically for conversational AI tutoring.
Voice-Native
Every text field has optional spoken variants optimized for TTS pronunciation.
Standards-Grounded
Maps to IEEE LOM, LRMI, SCORM, xAPI, QTI, CASE, and Open Badges.
Tutoring-First
Built for natural conversation with stopping points and misconception handling.
Unlimited Depth
Topics can nest to arbitrary depth for complex subject matter.
Server Components
UnaMentis includes a Python-based management server and web dashboard for monitoring, curriculum management, and analytics.
Management Server
Async HTTP server with WebSocket support for:
- Remote logging aggregation from iOS clients
- Real-time metrics streaming
- Resource monitoring (CPU, memory, thermal)
- Idle state management
Python 3.11+ / aiohttp / asyncio
Curriculum Database
Storage and retrieval for UMCF curricula:
- SQLite database with shared access
- Plugin-based import from MIT OCW, CK-12, and more
- Search and filtering by metadata
- AI enrichment pipeline for content processing
SQLite / Python / UMCF JSON
Operations Console
React-based administration interface:
- Curriculum Studio for viewing and editing content
- System health and service status monitoring
- Real-time metrics and performance tracking
- Plugin Manager for content source configuration
Next.js 16 / React 19 / TypeScript
Team
UnaMentis is developed and maintained by a dedicated team committed to making high-quality AI tutoring accessible to everyone.
Richard Amerman
Founder and Project Lead
Cy Goerdt
Partner