Elnino, a South Korean AI startup, has announced the official launch of Knoc, a real-time AI-powered translation subtitle service designed to make multilingual communication accessible to educational institutions, academic organizations, and international event organizers without the need for professional interpreters or costly equipment.

Conventional simultaneous interpretation typically requires trained interpreters, dedicated receiver hardware, and significant logistical coordination — often costing hundreds of dollars per hour and weeks of advance planning. For many universities, academic conferences, and mid-sized organizations, these costs put real-time multilingual support out of practical reach.

Elnino aims to change that equation. By replacing traditional interpretation infrastructure with a software-first, AI-driven platform, Knoc enables organizers to deliver multilingual experiences at a fraction of the cost and setup time. Starting from the Korean market — where international academic programs and cross-border conferences are rapidly growing — the company is expanding across Asia and beyond, targeting any environment where language barriers limit participation.

Stable subtitle output without retroactive corrections

Many existing real-time AI translation services process speech in short segments, displaying text immediately and then retroactively correcting it as more input arrives. This causes a “flickering” effect where on-screen text shifts repeatedly, increasing cognitive load for viewers.

Knoc takes a different approach. By leveraging a contextual memory window of up to 30 seconds, Knoc delivers subtitles that remain stable once displayed — producing output that flows naturally, similar to a live interpreter’s speech, without post-hoc corrections.

Flexible deployment across use cases

Knoc is designed to adapt to a range of environments:

• Classrooms and lectures: Instructors can use a smartphone as a microphone to start a session instantly. Where institutions previously needed dedicated interpretation hardware such as transmitters and receivers, Knoc replaces that infrastructure with a smartphone, significantly lowering the barrier to entry.
• Large-scale events: Organizers can connect Knoc to their existing audio systems (e.g., mixers). Elnino provides the software platform only; on-site operations run within the organizer’s existing staffing and logistics setup. Speakers require no additional technical configuration and can focus entirely on their presentations.
• Online meetings: Through Knoc’s web client screen-sharing feature, users can capture audio from video conferencing platforms such as Zoom or Google Meet — no additional hardware required.

Participants join via QR code or link, enter a room name in their browser, and select their preferred language — no app installation needed. Speaker language is detected automatically, and Knoc handles code-switching mid-presentation. Depending on the setup, a real-time Q&A chat feature is also available. A desktop overlay app for Windows and macOS allows subtitles to remain visible even during full-screen presentations.

Language support and data security

Knoc currently supports 101 languages. Specialized domain-boosting models for medical, legal, and IT fields enhance translation accuracy for technical terminology. After each session ends, all audio and translation data is deleted and is not used for AI model training — a policy Elnino refers to as Data Zero Retention.

Pricing

Individual users can access Knoc on a pay-as-you-go basis at $4.99 per session, with no contract required. Institutional and enterprise users can opt for custom plans that include multiple concurrent sessions, team member management, and shared credit pools.

A free 10-minute trial is available immediately upon sign-up at elnino.kr. For inquiries, contact: contact@elnino.kr.

Pilot results

Prior to the official launch, Knoc was deployed in pilot programs at university international programs and academic conferences in South Korea. The pilots covered multilingual classrooms, field trips, and conference sessions conducted in Korean, English, Japanese, and Chinese (both Simplified and Traditional). Among participants surveyed, 83.2% responded that continued adoption of the service was necessary.

“The era of needing separate budgets, equipment, and personnel just to enable multilingual communication should be behind us,” said Gyeongmin Kim, Founder of Elnino. “Language support shouldn’t require special preparation — it should be part of the default environment.”

About Elnino

Founded in 2024 as a student venture at Hanyang University, Elnino is a two-person AI startup focused on real-time multilingual communication. From its earliest days, the team has worked directly with end users by visiting university international programs and academic conferences to validate its product in real-world settings. Through multiple rounds of complete architectural redesign informed by user feedback, the team arrived at the current version of Knoc.