Rose Luk

Writer

LYM Founder

Profile

Join date: May 26, 2025

About

Rose Luk is the lead AI prompt engineer and founder of Lyone Media Group. She has 10 years of experience in songwriting and music production (DAWs, MIDI, VSTs). She also has over 5 years of experience in graphic design, UI/UX design, web design, business management, and marketing, and just under a year of experience in AI prompt engineering, filmmaking, and video production (Davinci Resolve, Capcut).

Her professional career began as an intern at the award-winning Cue Recording Studios in 2017. After receiving her Master of Business Administration in 2020, she founded music-entertainment startup Thaideke and did the UI/UX design for their independent music streaming service and online music store Sonicol. It was also during this time that she began her research into Christian theology, worldbuilding, conlangs, and epic poetry.

Rose made the decision to officially dissolve Thaideke in 2024 and head in another direction. The main factors in this decision were a Zoom conversation with PledgeMusic founder Benjie Rojers, the shutdown of Live Undiscovered Music (LUM), and the rise of generative AI.

In 2025, she founded Lyone Media Group and directed the company's first AI-short film "Vessel of Mourning" in 48 hours for the Runway Gen:48 competition using a variety of generative AI image, music, and video tools. Rose has experience prompting with Chat GPT, DAll-E, Soundraw, Udio, ElevenLabs, Runway, and Sora.

Posts (3)

Jul 11, 2025 ∙ 4 min

AI Speech-to-Speech Voice Conversion for Captain Vagor Using ElevenLabs: Demo 2

This article provides a breakdown of how I generated Captain Vagor's voice in our AI short film Vessel of Mourning (Gen:48 Edition)...

Jun 17, 2025 ∙ 4 min

AI Speech-to-Speech Voice Conversion for Captain Vagor Using ElevenLabs: Demo 1

As artificial intelligence-driven media workflows become increasingly common, transparency around how these tools are used is essential. In this post, we'll look at how I used ElevenLabs' Speech-to-Speech AI voice technology in one of our recent content marketing videos.

May 29, 2025 ∙ 3 min

Pushing the Boundaries of AI Video Generation: A Leap to One-Minute Narratives Using Test-Time Training

In the realm of AI-generated videos, creating coherent, long-form content has been a significant challenge. Traditional models often produce short clips, typically under 20 seconds, due to computational constraints and limitations in handling extended temporal contexts. However, a groundbreaking approach that will be presented at CVPR 2025 introduces a method to generate one-minute videos with coherent storylines using Test-Time Training (TTT).