Training Turk

Mercor listing

Sonic Audit Specialist - Italian

$40/hr

The role in one line

Audit AI training audio datasets in Italian, evaluating transcription quality and validating automated alignment boundaries.

Written by Training Turk from the public listing; it may be incomplete or out of date. Read the full posting on Mercor.

What you would do

  • Listen to recordings and compare against annotated transcriptions for accuracy
  • Judge whether word-level timestamps correctly align with spoken audio
  • Tag applicable error codes and assign pass or fail verdicts
  • Write detailed explanations for every decision in English
  • Identify acoustic ambiguity versus clear annotator mistakes

Who they are looking for

  • Native or near-native Italian speaker from Italy or 5+ years residence in Italian-speaking region
  • Strong English reading and writing for rubrics and feedback
  • Prior experience with transcription, subtitling, linguistic annotation, or QA work
  • Ability to distinguish similar sounds and identify word boundaries in speech
  • Disciplined approach to applying standards rather than relying on intuition

Skills this role asks for

audio transcription auditinglinguistic annotationphonetics and sound discriminationattention to detailrubric applicationerror classificationquality assuranceenglish written communication

What the interview is likely to probe

  1. 1.Transcription accuracy judgment

    The interviewer verifies you can reliably distinguish correct transcriptions from subtle errors, including regional accent variations.

    Expect something like: “You heard a dialectal pronunciation of a common word that the transcriber rendered incorrectly. Explain how you would determine whether this represents a genuine transcription error or acceptable variation.”

  2. 2.Word boundary identification

    Critical for forced-alignment auditing; you must identify segment boundaries accurately despite coarticulation and speech flow.

    Expect something like: “In continuous Italian speech, two words run together without clear pause. Describe how you would determine the correct segmentation boundary and when you'd flag a segment as misaligned.”

  3. 3.Rubric-driven consistency

    Quality requires applying written standards uniformly across hundreds of judgments rather than shifting standards based on context.

    Expect something like: “You encounter a borderline error that could fall under two different error codes. How would you decide which code applies, and how would you ensure consistency across similar cases?”

  4. 4.Written explanation clarity

    Subsequent team members depend on your rationales to understand decisions and validate quality; unclear explanations undermine the entire audit process.

    Expect something like: “You're explaining why a transcription failed. Your first draft mentions the error but doesn't explain why it matters. How would you revise your explanation to be clearer for someone unfamiliar with the audio?”

Exercise you may get

Review a sample transcription and forced-alignment segment, identify any errors, apply error codes, and write a brief rationale explaining your judgment.

How to prepare

  • Practice distinguishing similar Italian phonemes and regional pronunciation variations
  • Review sample transcription error codes and calibration examples
  • Familiarize yourself with waveform audio editing software basics
  • Prepare examples of audio QA or linguistic annotation work you've done previously

Facts

Pay
$40/hr
Commitment
hourly
Hours
10 per week
Work arrangement
remote · Remote
Posted
9/18/2026