A quick recap
What started in February as a weekend experiment has snowballed into more than fourty finished videos, two brand‑new characters and a community of families who now watch Mateo on their living‑room TVs. Along the way we’ve discovered practical ways to keep our clay‑animated hero on‑model, raised every episode to ultra‑sharp 4 K, and wrestled with the fine art of matching voices to moving mouths. Here’s the journey, stripped of studio slang.
From our first episode A to the end of April with P
Three months at a glance
- 22 full episodes (each about three minutes long)
- 14 short clips (half a minute each)
- Two new characters—Alphabitsy already has her own mini‑series in the works and another one yet to be announced.
- Picture quality jump from HD to full 4 K once we realised most families watch on big screens
Despite the extra pixels in 4K, a typical episode now reaches YouTube in almost the same time as our pilot thanks to a tighter workflow. Average episode production time is between 3-5 days.
Lessons we wish we’d had on day one
Keeping Mateo’s look steady Mateo is 3D, part clay figure and part plush toy. Early on we noticed that different image tools drew him with slight changes—an eyebrow moving here, a missing freckle there. Our fix was to store a small set of “golden” reference images and body poses outside any single software. Each time we generate a new scene, those references guide the model so Mateo’s face and body stay exactly the same. If you’re starting a series, pin down your character bible before episode one; changing the rules later is twice the work.
Voices that sit at the same loudness Different text‑to‑speech voices arrive at different volumes; some whisper, others boom. We now pass each voice track through software that measures how loud it sounds to the ear and nudges everything to the same point before mixing with music. The result: no more volume jumps when characters trade lines.
Lip‑sync: still the trickiest piece The most accurate mouth‑movement tool we’ve found is Hedra, but it only exports regular HD, not 4 K. We now create the mouth movements in HD, enlarge the footage with separate software, and blend it back into our 4‑K scene. It works, but it adds an extra step.
Brands and on‑screen words finally stay sharp Back in March one of our favourite models quietly improved the way it draws text. Suddenly titles and logos stopped looking melted. We can now drop brand names—or the letter of the day—straight into a scene and trust the spacing and shape will hold. Small fix, big relief.
Shorts that feel born for vertical screens Our early short clips looked like cropped versions of the main show. After a dozen tests, we relaunched as Llama Shorts and AlphaBitsy Shorts in late April. They tell quick jokes, stick to a brisk rhythm and export in under two hours. Viewers now finish these bite‑sized stories 22 per‑cent more often than full episodes.
What’s next
- Finishing the alphabet: 26 main episodes plus 26 shorts ready by mid‑June.
- Alphabitsy’s YouTube channel: Launching in early June.
- Mateo and The Llama Alphabet Album: launching on all music streaming platforms this week.
- Subscriber‑only webinar: If time and demand allow us we’ll open our project files and walk through clay, puppet, 2‑D and 3‑D workflows. Newsletter readers get first access.
Follow the adventure
Every new episode (and the occasional blooper reel) drops first on YouTube: Mateo & The Llama. Let us know which letter, llama joke or wardrobe swap makes your day. Every Subscribers helps, so when over at Youtube don’t forget to hit subscribe and notifications.
Thanks for reading—and for cheering on a clay‑animated llama built entirely with generative tools. If these notes helped, hit “Subscribe” so you don’t miss the webinar invite or the next behind‑the‑scenes dispatch.
See you after the next render, The Storyplai Team
