Show HN: Worklaude – learn by speaking

Not documented. The post says each step is a short video where the user can speak and interact at the end, and nothing more. No question, branch, or threshold is named, so the branching logic cannot be reviewed from the thread.

A prototype, by the author's own description. The post says it is a small prototype for learning by speaking with video-based scenarios, still early and looking for feedback. The linked entry URL is the access page, where you enter an email to receive a link.

Only routing is documented. Whether pronunciation, grammar, or fluency are assessed, and what feedback you get back, appears nowhere in the post or the access page, which lists sequences and scene counts and nothing about scoring.

Nothing published says more than English. The access page lists workplace and travel sequences, a job interview, a pharmacy visit, buying a train ticket, hotel check-in, and an allergy order, five scenes each, with no language selector or level progression documented. The author did not respond in the thread.

Almost none. The captured page shows 1 point, posted by the handle j_rcht, with no comments on it and no author reply in the capture. The only extra text is a help link, so the thread is a pointer to the prototype rather than a discussion of it.

That the video scenes branch on TypeSafe JEV. The body says only that each step is a short video where at the end the user can speak and interact, still early and looking for feedback. No scene question, option set, branch condition, or threshold appears anywhere in the capture.

None. The captured page gives no pricing, no latency, no scene count and no user count, and names nothing beyond JEV in the title. Anything quantitative would have to come from the product itself rather than from this thread.