What the work involves

You work through queues of short Italian audio clips and score them on dimensions like nativeness, fluency, pronunciation accuracy, and prosody. Some clips are recordings of human speakers; others are synthetic speech produced by a model being evaluated. Each rating needs a written justification in English that points to specific evidence — a devoiced final consonant, a misplaced stress in a paroxytone, intonation that reads as read-aloud rather than conversational, a regional feature that is authentic but may or may not fall inside the rubric's definition of standard.

The hard part is not hearing the difference. It is describing it in a way another reviewer would reproduce. Expect calibration rounds, occasional guideline updates mid-project, and back-and-forth with coordinators when clips sit in genuinely ambiguous territory — for example, a fluent L2 speaker with near-native segmental accuracy but non-native rhythm.

What the screen looks for

micro1 runs an AI-led interview before任 project placement. It tends to probe:

  • Verifiable Italian background — where you acquired the language, what register and region you're strongest in, and whether you can discuss variation (Tuscan vs. Roman vs. Northern realizations) without hedging.
  • Phonetic vocabulary under follow-up — you don't need an IPA degree, but if you say "the intonation was off," expect to be asked which contour and where.
  • English writing quality — B2 is the floor; the justifications are the deliverable, and vague ones get flagged.
  • Rubric discipline — whether you can hold to a written standard you personally disagree with, and flag it separately rather than scoring around it.

Logistics

Remote, contractor, asynchronous. Work is assigned in batches with deadlines rather than fixed shifts, so you set your own hours; most contributors report 10–25 hours a week when a project is active, with gaps between engagements. You'll need quiet listening conditions and decent over-ear headphones — laptop speakers are not adequate for judging fine-grained pronunciation. Pay in the $30–65/hr range has been observed on similar micro1 language engagements; the actual rate is set per project and is not guaranteed.