AllComputer GigsCreative GigsCrew GigsDomestic GigsEvent GigsLabor GigsTalent GigsWriting Gigs
Location
Remote only
Estimated duration
~1h
Applications close
USD
We're collecting conversational audio in Sesotho for a global AI training dataset. We're looking for native Sesotho speakers to record and submit audio directly.
What you'll do:
Recording mix across your submissions:
What you'll be paid:
Quality standards:
Timeline: Rolling submissions Jul 20 – Sep 14, 2026.
What to submit per audio hour:
Rejected audio is not paid and cannot be resubmitted. Please review the quality standards carefully before recording.
One submission = one accepted audio hour in Sesotho, meeting all technical quality standards, delivered with metadata and (on first submission) signed Speaker Consent Form.
EVIDENCE DETAILS
Link: URL to audio file(s) totaling one accepted hour. Text: full metadata (Speaker ID, age, gender, speaker count, length, sample rate, bit depth, channels, format). Photo: signed Speaker Consent Form on first submission.