Skip to main content
InterpScout is a teaching agent for interpreter training. An agent, here, is not a chatbot that answers questions — it is software that does the whole job, end to end. Ask a chatbot to prepare a course and it tells the teacher how; InterpScout’s agent actually produces the finished course — sourced, aligned sentence by sentence, glossed, and machine-checked — in minutes. Students then practice against it aloud while a model scores every attempt and routes the results back to the teacher. It is not a demo. InterpScout runs in production at interpscout.org, used by a practicing interpreting teacher and her students at Sichuan University. Every artifact on this page was produced by the live system — nothing here is a mockup.
评委您好 — 如果您想先用中文快速了解产品逻辑与测试路径,请直接阅读 评委导览(中文)。该页已覆盖评审所需的全部信息;英文正文为技术评委提供更深细节,非必读。

See it for yourself

Three real artifacts, pulled from the running database — this is what “an agent does the job” means in practice.

A graded glossary term

Every term the Producer curates carries a grade and a source. A rendering is marked 官方译法 (official) only when the agent actually found a citable bilingual source — never upgraded on confidence alone.

A real scored practice attempt

A student interpreted a sentence aloud; a multimodal model listened to the audio and scored it on three dimensions, with reasons.Source (中 → 英): 当前正在发生的以人工智能为标志的科技革命,将对新一轮经济全球化和人类社会发展产生难以估量的深刻影响。What the student said: “The ongoing scientific and technological revolution characterized by artificial intelligence would exert an unimaginable impact on the new round of economic globalization and the development of human society.”The model’s note:unimaginable impact 略异于更标准的 profound impact,但意思相近;exert an impact 是地道搭配;术语正确:人工智能、经济全球化。”When a student misses an official term, it is routed automatically into that student’s spaced-repetition flashcard queue.

A briefing's bilingual term table

Part of a real 9,000-character pre-meeting briefing (WTO MC14). Each row is graded and sourced, ready for an interpreter to prepare against.

In numbers

33 real courses

Produced from verified bilingual sources — e.g. the 2026 Government Work Report: 218 terms, 246 aligned sentence pairs.

9,000-character briefings

Pre-meeting knowledge packs with graded bilingual term tables and cited further reading.

3-dimension scoring

Multimodal audio scoring — accuracy · terminology · fluency — with automatic weak-term flashcards.

Live in production

On Vercel with a durable agent runtime, used by a real teacher and her students.

What InterpScout does

An interpreting classroom has three chronic bottlenecks: preparing bilingual teaching material is slow, pre-meeting preparation is unteachable at scale, and practice feedback depends entirely on teacher hours. InterpScout assigns an agent to each.

Producer — course production

The teacher points at a bilingual source (or picks from daily auto-discovered candidates). The Producer agent aligns it paragraph by paragraph, curates terminology across the whole document, and publishes a parallel-reading course — typically in minutes.

Briefing — pre-meeting preparation

Given a meeting topic, the Briefing agent researches the web, cross-checks sources, and delivers a 9,000-character pre-meeting knowledge briefing: domain crash course, speaker and institution profiles, contested issues, and a graded bilingual term table.

Practice — sight interpretation

Students interpret real sentences from real courses, out loud. A multimodal model listens to the audio directly and scores accuracy, terminology, and fluency — then routes every missed term into that student’s flashcard queue.

Classroom — management & analytics

Teachers create classes with invite codes, watch per-student averages, and see the terms the whole class keeps missing — data that decides what next week’s lesson should be.

Start here

1

Read the Chinese judges guide

评委导览 covers the whole product logic and the test-account path in Chinese — the fastest way in.
2

Log in and try it

The Quickstart walks you through the live site with ready-to-use accounts — produce a course, generate a briefing, score a practice attempt.
3

Watch an agent work

Every course and briefing reports progress live on the Activity stream — you can watch the research desk fill in, in real time.
4

See how it's verified

Truth Layer explains the core thesis: a prompt is never a guarantee — a machine checks what a machine produces.

How the pieces connect

The InterpScout loop: a teacher submits a source or meeting topic to the Producer and Briefing agents; their output and a shared terminology base flow through student practice, AI scoring, flashcards, and dashboards, and back to the teacher.

The teaching loop: a teacher hands a source or a meeting topic to the two agents; their output and a shared terminology base flow through student practice, AI scoring, flashcards and dashboards, and back to the teacher.

Terracotta marks the two autonomous agents; gold marks the shared terminology base that every course and briefing both feeds and draws from. What connects them is a cross-document terminology base: every course a teacher produces deposits its graded, sourced terms into a shared, searchable glossary that gets thicker with use — so the thirty-first course on trade policy is better sourced than the first. See Terminology base for how that shared store works.

Why agents, not a chatbot

A chatbot answers questions; a classroom needs artifacts — a finished course, a finished briefing, a finished score. InterpScout’s agents run as durable, long-lived sessions that survive deployment restarts, report progress in real time, and are held to mechanical quality gates: a course cannot be saved with misaligned paragraphs; a briefing cannot be saved without its term table and verified further-reading links. When a gate rejects, the agent reads the reason and revises inside the same session. The teacher sees only the finished artifact, or a clear explanation of what a source could not support. Read Truth Layer for how deep that principle runs, or Philosophy for why it is anchored to a real classroom.