Docs / source reviewed
LinguaTrace
Turn language-lesson corrections and vocabulary into revision notes
How this project uses Jev
- Input
- Speaker-labeled lesson transcripts, recent turns and candidate spans from the original text
- Jev decides
- Whether a turn is lesson content, contains a learner error or tutor correction, and which spans to quote
- Code executes
- An external speech service transcribes audio; code pairs corrections with learner utterances and assembles notes, flashcards and gap-fill exercises
Evidence and limitations
Jev does not hear audio or replace the tutor. Transcription errors can produce incorrect notes, and fixed demo data is not evidence of live-model quality. We did not run a lesson or evaluate learning outcomes.
This project has not been run independently here. Author-reported results are not independently verified results.