Long-term, part-time contract for native Korean speakers based in South Korea to transcribe short audio clips; flexible ~1 hour/day schedule at $14/hour. Ideal for detail-oriented listeners with strong written Korean and steady internet.
Audio & Speech
100% Remote Hourly · $14/hr
$14/hr
Compensation
Worldwide
Eligibility
Entry
Experience
Jan 7, 2026
Posted
Open worldwide
Interested in this role?
Create a free OpenTrain account and apply in minutes.
OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect contributors to meaningful, long-term projects where human judgment helps shape how AI understands and uses language.
Creating an OpenTrain account is free; we hire and contract directly for this role and provide the platform, guidelines, and tasks you’ll use every day.
About AI training (data labeling) work
AI training is the human side of building intelligent systems: people prepare, check, and correct examples that models learn from. Tasks include transcription, reviewing audio, labeling speech segments, and applying clear rules consistently.
This work is 100% remote, often flexible and part-time, and a fast-growing way to work in tech—contributors directly influence how speech-enabled systems perform.
The role at a glance
We’re hiring native Korean speakers located in South Korea for a long-term transcription and audio review contract. You’ll work roughly 1 hour per weekday with a flexible daily cadence.
This is contract, part-time work paid at $14 USD per hour. Accuracy and consistency are more important than speed.
Location: South Korea only (applicants must be currently located in South Korea)
Schedule: ~1 hour/day, Monday–Friday; flexible timing each day
Duration: Long-term (12+ months)
Employment: Contractor, part-time
Pay: $14 USD per hour
Data: Short Korean audio files — transcription and audio verification
What you'll do day-to-day
You will listen to short Korean recordings and produce clean, standardized transcripts that reflect what was said. This is transcription, not translation or creative writing.
You’ll also perform simple listening-based verification and labeling steps using the platform’s predefined options—flagging unclear audio, marking unusable segments, and following strict conventions for uncertain passages.
Transcribe short Korean audio clips into accurate, standardized Korean text
Apply platform rules and formatting consistently (spelling, spacing, punctuation)
Flag unclear or unusable audio and mark segments you cannot confidently transcribe
Make careful judgment calls for imperfect audio (minor mispronunciation vs. different words)
Prioritize consistency and accuracy over speed
Requirements
You must meet all location and language requirements and have the equipment and connectivity to work reliably each day.
Must be currently located in South Korea (South Korea–based applicants only)
Native Korean speaker with native-level written accuracy
Strong command of Korean spelling, spacing, punctuation, and standard writing conventions
Comfortable following strict guidelines and applying platform conventions without improvisation
Able to handle imperfect audio and mark unclear segments appropriately
Able to make careful listening judgments about similar-sounding words and pronunciations
Commitment to ~1 hour/day, Monday–Friday, long-term (12+ months)
Reliable internet, a computer suitable for web-based tools, quality headphones, and a quiet environment
Willing to complete a short Korean transcription screening sample before starting
Prior transcription, captioning/subtitling, or speech labeling/QA experience preferred but not required
Who should apply
This role is well suited to careful, patient people who enjoy focused listening and applying clear rules. Entry-level applicants are welcome if you meet the language and location requirements.
If you take pride in accurate, standardized Korean text, can work reliably each weekday, and prefer a steady, part-time contract that fits around other commitments, this role is a good fit.
Join OpenTrain to evaluate AI-generated Korean transcriptions and diarization for multi-channel conversational audio; remote, part-time contractor work requiring 20+ hours/week and strong Korean transcription/QC skills.
Join OpenTrain to review multi-channel Korean audio, verbatim transcripts, timestamps, and diarization for high-accuracy multilingual AI evaluation data; native Korean required, 20+ hours/week, part-time contractor, fully remote.
OpenTrain AI seeks native Japanese speakers in Japan to transcribe short audio clips and verify audio quality; long-term, flexible schedule (~1 hour/day, Mon–Fri) at $15/hr. Must follow strict guidelines and complete a short screening sample.