Our client is building first-person (POV) video datasets used to train AI models that learn to connect what they see (objects, hands, and actions) with what they hear. The footage shows real people performing hands-on tasks, recorded from a head- or chest-mounted phone.
We're looking for English-fluent narrators to watch this pre-recorded footage and describe the scene and task in the first person, as if explaining it to someone who can't see. You don't need to know how to do the activity on screen. What matters most is your fluency and your ability to describe clearly, continuously, and accurately what's happening.
Videos span a wide range of everyday and specialized activities, including home and daily tasks (cleaning, laundry, pet care, packing), repairs and DIY (furniture assembly, plumbing, bicycle maintenance, soldering), textile and craft arts (sewing, knitting, pottery, leathercraft), professional trades (automotive repair, warehousing, lab work, assembly-line packaging), hobbies and arts (painting, instruments, model building, camping), technology (devices, gaming, VR/AR,
3D printer setup), outdoor activity (city tours, shopping, public transit), personal care (haircuts, makeup, grooming), and specialized professional work (first aid, barista workflows, chef knife work, lab protocols).
What You'll Do
• Watch pre-recorded POV videos and record a live, first-person English narration of the setting and each task step
• Deliver audio narration synced to the video footage, recorded after the footage was captured
• Maintain a steady narration density of at least 25 words per minute on average, with no long silences
• Keep 90% or more of your narration focused on the setting or the task steps being performed
• Follow the project's Acceptance Criteria, which will be shared in full before work begins
• Complete an initial test video; approval of the test video unlocks immediate start
What You Bring
• English fluency — this is the #1 selection criterion
• Ability
📌 Video Narrator (Modena)
🏢 Toptal
📍 Modena