Voice & Audio AI Trainer - Dubai

Role Overview:
Join an AI research project as a Voice & Audio AI Trainer and help build high-quality spoken datasets for multimodal models. You’ll create natural, descriptive audio that narrates visual content — a key input that helps train systems to understand both images and speech.
Key Responsibilities:
• View images and craft concise, natural spoken descriptions that accurately convey visual content.
• Record short audio clips (typically ~2–3 minutes each) using provided tools or platforms.
• Ensure high audio quality — recordings should be free from background noise, distortion, or interruptions.
• Follow precise linguistic, timing, and stylistic guidelines supplied by the research team.
• Work with QA teams to review recordings and incorporate feedback to improve dataset quality.
Qualifications:
• Excellent verbal communication and clear enunciation.
• Native or near-native fluency in English (additional language skills are a plus).
• High attention to detail and strong ability to follow annotation rules exactly.
• Prior experience with voice recording or annotation is helpful but not required.
• Comfortable working independently and maintaining consistency when performing repetitive tasks.
Project Details:
• Schedule: Fully remote and flexible — fit tasks around your availability.
• Commitment: Part-time, self-paced; work is assigned in batches.
• Duration: Ongoing, with periodic opportunities as new batches are released.
Compensation & Contract:
• Pay Rate: $21/hour (USD).
• Engagement: Independent-contractor arrangement.
• Payments: Weekly, processed via Stripe Connect.
Application Process:
• Submit your resume and availability.
• Complete a short AI interview (~15 minutes).
• Fill out a brief scheduling form.
• Typical response time: Applicants usually hear back within one week.
APPLY HERE
https://gulfcareerhunt.com/hiring-voice-audio-ai-trainer-remote-usa/

Posted on : 3 weeks ago, #205765, 3 views

It is ok to contact this poster with commercial interests.

← Previous Next →