Voice-first interface
Just talk. Whisper on-device ASR transcribes your explanation in real time — no typing, no upload, no latency.
Speak out what you know and let AI surface exactly what you still need to learn — using the same method Nobel Prize winners swear by.
The Feynman Technique is one of the most effective learning strategies ever documented. Teachback turns it into a 10-minute daily habit.
Choose what you want to learn — a concept, a chapter, a skill. Teachback suggests review topics based on your history.
Explain the topic as if teaching a beginner. Your voice is transcribed on-device in real time using Whisper AI.
The AI asks probing questions about anything you glossed over or got wrong — pinpointing exactly where your understanding breaks down.
Each session generates a four-dimension mastery report. Spaced-repetition scheduling brings weak concepts back at the right time.
Just talk. Whisper on-device ASR transcribes your explanation in real time — no typing, no upload, no latency.
Questions are tailored to your subject (STEM, language, humanities) and escalate in depth as your mastery grows.
Four-dimension scoring: accuracy, depth, coherence, and breadth. A circular progress chart shows your trajectory over time.
Teachback automatically schedules concept reviews at optimal intervals so knowledge stays in long-term memory.
Attach diagrams, textbook pages, or whiteboard photos so the AI asks questions grounded in your exact study material.
Use OpenAI, Anthropic, or Gemini directly by entering your own API key. Or run fully offline with a local Ollama model.
Transcription happens locally using Apple's CoreML-accelerated Whisper model. Your recordings are never uploaded to any server.
Every plan includes unlimited local sessions. Cloud AI sessions are shared with a generous free quota.
Free
$0
Perfect for getting started
Gemini Flash LiteStandard
$4.99 /mo
Serious learners
Gemini FlashPro
$9.99 /mo
Power users
GPT-4oUltimate
$14.99 /mo
Maximum intelligence
Claude