English Chinese Bilingual Voice Recording AI Trainer

About the Role

Mindrift Data Annotation Projects is seeking an English Chinese Bilingual Voice Recording AI Trainer for a part-time, project-based opportunity in Dammam, Eastern region. This role involves recording natural, conversational voice clips that blend English and Chinese, contributing to an AI training platform. There are no fixed hours or commitments beyond what aligns with the individual's schedule.

Role Context

This position focuses on capturing authentic bilingual speech patterns for AI model development. Candidates will engage in unscripted recordings, discussing everyday topics, opinions, or thoughts, naturally switching between English and Chinese as they would in real-life conversations. The primary goal is to provide diverse and realistic linguistic data.

Responsibilities

• Record voice clips, typically 1-2 minutes in length, in response to prompts or short videos.
• Speak naturally, incorporating both English and Chinese as would occur in daily conversation.
• Complete multiple clips per recording session.

Qualifications and Requirements

• Language: Fully bilingual, fluent or native in both English and Chinese, with a natural tendency to mix the two languages when speaking.
• Experience: 0-1 years of relevant experience.
• Equipment: Any recording device, including a smartphone. No specialized studio equipment is required.
• Recording Environment: A quiet room free from background noise such as music, television, traffic, or other voices.
• Audio Quality: Clear voice, recorded close to the microphone, with only one speaker and no echo. A headset or earbuds with a microphone are beneficial but not mandatory. Clean audio is critical for approval.

Compensation

Compensation is performance-based, with individuals earning up to $30 per hour, dependent on recording pace and approval rate of submitted clips.

Work Arrangement

This is a flexible, remote, part-time opportunity where individuals are compensated per approved recording. The work can be performed from any comfortable location, utilizing standard personal devices.