Before you start
A coding agent, browser and three phrases verified by a fluent speaker or trusted source.
Why this lesson exists
This lab adapts the “Build a 3D Language-Learning RPG” project write-up into a practice build. The time is an estimated first session, not a promise to finish a production system. Use the public repository as a reference when available; the exercise can be built with original sample content.
Do the exercise
Define the practice version
Keep lesson data separate from the scene. Start with one location and a single question before adding voice or 3D polish.
Build step 1
Write one scenario: enter a hotel and say hello.
Build step 2
Store the target phrase, translation, hint, and accepted variants as data.
Build step 3
Build a tap-to-continue version first.
Build step 4
Add recorded example audio.
Build step 5
Add speech recognition and confidence feedback.
Build step 6
Add a 3D environment only after the lesson works as plain text.
Run the experiment
Test correct, incorrect and missing answers. Deny microphone access and finish the encounter using tap or text. Then ask a new transfer question.
A prompt to adapt
Replace the bracketed parts with your own practice details.
Work in a disposable practice project. Explain any setup requirements before changing files. Build one small step at a time and show how I can check it. Design one five-minute survival-language scenario for a beginner checking into a hotel. Return a JSON schema with scene goal, target phrases, translations, pronunciation hints, accepted variants, failure hints, and completion condition. Review this speech-learning interaction for accessibility. It must work for a learner with no microphone, an accent the recognizer struggles with, slow reading speed, and intermittent internet. My first-version boundary: Keep lesson data separate from the scene. Start with one location and a single question before adding voice or 3D polish.
Run this experiment
Test correct, incorrect and missing answers. Deny microphone access and finish the encounter using tap or text. Then ask a new transfer question.
Check your result
Use evidence from your output. A confident explanation from the AI is not enough.
- Lesson content can be reviewed without reading rendering code.
- An incorrect answer gives feedback and a retry.
- Voice failure does not block progression.
If it isn’t working
Do not confuse speech-to-text success with pronunciation assessment. Verify both game state and learning separately.
Optional. Progress stays in this browser.
Where this came from
Public project repository ↗. The practice lesson is an adaptation, not a verbatim transcript. About the sources.
Prepared September 2026. Tools and interfaces change; use current official setup instructions. Session lengths are estimates.