You must be native-level in French and comfortable producing or validating ARPABET phonemic transcriptions at an intermediate to expert level. You should be able to reliably detect reading errors and speech disfluencies with high accuracy. Backgrounds in phonetics/linguistics, ASR datasets, or child-speech transcription are strongly preferred. You must be able to start within one week, pass a qualification quiz, and commit 20+ hours per week at the published rate. You will convert large batches of long (1–2 minute) French audio clips into precise word- and phoneme-level transcripts using our web-based annotation tool. Work is delivered in rolling batches, and your annotations will go through a peer QA loop before final acceptance. This work will directly support next-generation speech recognition and language-learning models. Payment is per approved task—higher throughput and accuracy mean higher earnings.
Estimated Total Earnings
$1.32
Pay per Hour
$22.00/hr
Time Requirement
20+ hrs/week
Duration
3-6 months
French audio clips needing phoneme transcripts
Software
Hiring Type
Required Location
Workload / Schedule
20 hours per week, Flexible and able to start immediately
Software
Data Type
Task Types
Subject Matter / Industry
Language
Job Type
Proposals: 15
Invites sent: 0
Unanswered invites: 0
Share link