Varco Sound accepts images and video as well as text. Users can describe a situation in a sentence — such as “water droplets in a dark cave” — or upload a forest photo or battle footage, and the service analyzes the scene’s context to automatically generate sound files, or sound assets.
The generated audio is provided in “multitrack,” separating sound elements into individual tracks. That allows detailed postproduction edits, such as adjusting the volume of a specific sound or swapping out an element. NC AI said the goal is to reduce the “difficulty of re-editing” often cited with single-track audio generation.
The service also emphasizes variation. It can quickly produce multiple versions while keeping the texture and nuance of a user-uploaded sample, expanding options and reducing prototyping time.
Varco Sound runs on the web without a separate installation. Users can sign up with email verification, and it is offered through the Varco Game Package plan, which combines it with the 3D generation AI service Varco 3D. Pricing includes a 22,000-won-a-month Plus plan (10,000 credits) and a 110,000-won-a-month Premium plan (50,000 credits). New users receive 2,000 credits.
NC AI said it aims to refine quality by incorporating feedback during the beta period and target an official launch in the first half of this year. The company is also developing plugins for game engines Unity and Unreal, as well as plugins that run directly in professional audio production software.
“Varco Sound is a technology that turns imagined sounds into assets that can be produced immediately,” NC AI CEO Lee Yeon Su said. “Based on our multimodal AI capabilities, we will expand it into a platform that leads the global content production market.”
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.