Hello there, my name is Jean-Bernard Lauture. I am an aspiring author, writer. I've got a background in Music, editing and Data analytics. AI has allowed me to bring my stories and ideas to life in ways I never thought possible. Here to learn, share what I know and grow with you all.

Got a year of Premium mostly to see if I can actually lipsync in it. If Kavan the Kid can, it must be possible.

WARNING: Contains themes of psychological horror, violence and harsh language. This was not supposed to be a full video project, however once I started I couldn’t stop working on it. I just wanted to play around and make something to release a bit of stress. It ended up becoming a much more deeper exploration into the topic of mental health. Specifically dealing with battling depression, isolation and loneliness. Struggles I face in my own life, an ongoing inner struggle being torn between fear, anger and sadness. At times feeling overwhelmed by the darker side. This is my attempt at trying to visualize how it feels not being seen or understood while at the same time fighting personal inner demons. On the tech side, because I didn’t want to invest too much into this project at the start. I challenged myself to use only free tools. Generating a few free videos a day. This of course makes the process slower, but that is how I like it. Gives me time to think and really dive into the story that I want to express. I used the following: Google Flow mainly for image creation. Some video shots with VEO. Most of the lip-sync was done with Wan 2.7 1 or 2 Seedance 2 shots. A few other Seedance 1.5 shots. Suno for the music. Eleven labs for a few sound effects. All the above have some kind of free level that gives a few free generations each day. Post production, with Adobe photoshop and final edit with premiere.

Had an idea so I sent it to a developer. Wish me luck :) if you are interested in, let me know, no deal has been struck yet. Character Voice Locking System (CVLS) — A Feature Proposal for Resemble AI Submitted by: Jonathan Wilson / Studio of Worlds Category: Product Feature Proposal Date: June 2026 The Problem AI voice generation has solved quality. It has not solved consistency. Creators working in long-form narrative — animation, serialized storytelling, games, film — need characters who sound the same across hundreds of scenes, across months of production, across the full emotional range of a performance. Currently, achieving this requires rebuilding voice parameters from memory on every session, with no systematic way to lock, save, or reuse a complete character voice profile. The image generation community solved this problem with model sheets and character ICILs (Image Consistency and Identity Lock). Voice generation needs the same solution. The Proposal: CVLS — Character Voice Locking System A saved, reusable character voice profile that functions as the audio equivalent of a model sheet. Layer 1 — Base Voice Profile (The Lock) A permanent character record containing: Voice ID (selected from Resemble AI library or cloned) Gender Approximate age Ethnicity / cultural background Accent and regional dialect Baseline speaking pace Baseline pitch and tone descriptor Character notes (e.g. "raspy from years of smoke", "formal register", "never raises voice") This profile is saved, named, and recalled instantly. It is the character. It does not drift. Layer 2 — Emotional Spectrum Mapping A pre-defined set of emotional states mapped to the locked voice, each with adjustable intensity on a 1–10 scale: Neutral (baseline) Joy / Happiness Anger Rage Sadness / Grief Fear Disgust Surprise Tenderness Laughter Crying Whispering Shouting / Yelling Sarcasm Exhaustion Each state is not a binary switch but a dial. "Anger at 3" reads as irritation. "Anger at 9" reads as barely controlled fury. The creator selects the state and the intensity before generating.

This all started June 16th 2025, it is funny to think on how much difference a year can make, I knew nothing of AI a year ago, I have not taken a day off since. I do not exaggerate, now one day, I hope to create full films just like this intro (hopefully better). I still have much to learn. Crystal is still a very important character central to this story but I wanted a shorter name and Magma Lad is the main character. Very happy with Mag's flaming hair, the environment and the logo treatment. It is all rendered in Kling 3.0 in 4K. 60 credits per 10 seconds. I learned it is essential to create your environments separately and then insert/assimilate your characters into your scene with Nano Banana Pro, it is how I got my starting frame. Then I used Nano to create my logo and used Kling to animate it. Eleven Labs made me a 20 second music track and all the SFX were made in Kling. Popped it all into After Effects and done.