Korean Audio Recorder and Transcriber

The work

You will help prepare Korean audio data for speech and language AI models. The work combines audio recording, transcription, evaluation, and annotation.

• Listen to Korean audio clips and write accurate transcriptions or evaluate their content.
• Record yourself reading provided sentences in Korean from a stationary car or quiet office.
• Create requested background sounds, such as seatbelt clicks or window movements, during recordings.
• Review and annotate audio data using detailed project instructions.
• Submit recordings and written feedback through the project platform.
• Follow privacy and data-use requirements for research and system improvement.

What it pays and takes

This is a part-time contractor role for an entry-level contributor. Previous AI training experience is not required, but you must be able to work independently and meet project deadlines.

• Pay: $20-$65 per hour.
• Time: 20+ hours per week.
• Language: Native-level Korean and at least B2-level English for instructions and basic feedback.
• Location: Open to applicants in the countries listed for this role.
• You need strong attention to audio quality, Korean accuracy, and annotation details.
• You should be comfortable using digital platforms and audio tools remotely.
• You must commit to privacy and responsible data handling.

How it works

Apply on OpenTrain with your resume and then complete the application on the hiring site.

About AI training work

AI training is the human work behind systems that understand speech and language. People transcribe, record, review, and evaluate examples so AI models can learn from accurate real-world data.