Work · Video, animation and learning

AI voice, music and media generation library

98 logged, reproducible generation jobs; cloned voices in English and Swahili; original music.

Status
BuiltLiving library, Jun–Aug 2026
Period
2026
Role
Author
Brand
Tuchopini
AI voice, music and media generation library screenshot

Every fal.ai job lives in a dated folder with a prompt.md recording model, parameters, prompt and outputs. Custom voice models (GPT-SoVITS, Applio/RVC, MiniMax) for character narration including Swahili; an Arcane-style video-to-video restyle in parallel chunks; client jobs such as the Xynergym logo reveal and Kenya Homes aerial imagery; 16 Suno songs with stems; an Ambush Tunes gospel/Afro catalogue produced in FL Studio.

Reproducibility as discipline: anyone can re-run any asset from its logged prompt.

Watch

Xynergym logo reveal (5 s) · 0:05
Arcane-style video-to-video restyle sample · 0:04
Forgiveness Unbound, Manim animation (42 s) · 0:42
Awards Night motion flyer · 0:05
"Sing" by Ambush Tunes (74 s) · 1:15
Morning Hate animation (53 s) · 0:54
Dance: "I'm just a boy on the inside" (59 s) · 1:00
Breakdance choreography to an AI beat (2023) · 0:37

Listen

  • Nostalgia (Ambush Tunes, FL Studio, 2026)

AI voice, music and media generation library image
AI voice, music and media generation library image

More video, animation and learning