Vega Audio Data Collection, United States

OneForma is seeking bilingual speakers to participate in the  Vega Audio Data Collection project. 

In this remote, task-based project, participants will record 6–7 realistic dialogue sets based on provided scripts. Each dialogue simulates a natural conversation between a customer service agent and a customer. 

Your role is to bring the script to life with fluent, conversational delivery, naturally alternating between two languages (code-switching) as indicated. Recordings must be clear, high-quality, and completed using a quiet recording environment and reliable equipment. 

This is a one-time task that takes approximately 1–2 hours to complete. Compensation is provided after successful quality review of submitted recordings. 

The goal of this project is to collect high-quality, natural conversational speech data to support the development and improvement of voice technology systems. A key focus is authentic code-switching — the natural practice of alternating between two languages within a conversation. 

• Native speaker of the main project language (apply to see the list) 

• Upper-Intermediate to Proficient (L2) level in the secondary language (apply to see the list) 

• Ability to naturally alternate between both languages in conversation 

• 18 years of age or older 

• Access to a reliable computer and high-speed internet 

• Professional-quality microphone or headset 

• Ability to record in a quiet, private environment without background noise 

• Detail-oriented and able to follow script instructions carefully