Audio Model Trainer

Hourly Contract | Part-Time Remote

Compensation: $20/hour

Role Overview

The Audio Model Trainer role supports a large-scale AI research project focused on developing multimodal systems that understand both visual and auditory content . Trainers record and evaluate short audio clips describing visual inputs, helping to create datasets that drive the next generation of AI models.

Key Responsibilities

• View video clips and provide preference-based evaluations.
• Record audio clips (2–3 minutes each) using project-provided tools.
• Ensure recordings are consistent, clear, and free from background noise.
• Follow linguistic, timing, and formatting guidelines precisely.
• Collaborate with research and QA teams to ensure high data quality.

Qualifications

• Strong verbal communication with clear enunciation.
• Native or near-native English fluency (additional languages are a plus).
• High attention to detail and ability to follow structured instructions.
• Prior experience with voice recording or data annotation is a plus, but not required.
• Reliable and comfortable working independently on repetitive tasks.

What You’ll Gain

• Contribute directly to foundational AI research .
• Gain experience at the intersection of language, audio, and computer vision .
• Enjoy flexible, fully remote work.

Compensation & Process

• Rate: $20/hour.
• Application Process:
• Complete a short AI interview (~15 minutes).
• Fill out a quick availability form.