From UX Designer to AI App Builder: My Journey into Creating an AI-Powered Kids' Storytelling App
I have watched several YouTube tutorials about creating kids' channels and animated videos, and I noticed how easily people are now creating animated content using AI tools. With just a prompt, these tools can take care
I have watched several YouTube tutorials about creating kids' channels and animated videos, and I noticed how easily people are now creating animated content using AI tools. With just a prompt, these tools can take care of everything from storyboarding and scriptwriting to suggesting ideas and generating voiceovers.
I became curious about how this process actually works, so I decided to experiment with it myself and explore its pros and cons.
As a UX designer, I wanted to take this a step further. Instead of relying entirely on third-party tools, I decided to explore how I could build my own application to create animated videos. My goal is to understand the technology, experiment with AI capabilities, and build a customized tool that gives me more control over the entire video creation process.
In this article, I'll share my learning journey, the challenges I encountered, and my experience building an AI-powered animated video creation application.
What I discovered from my research?
After watching plenty of YouTube videos, I noticed that many creators are already using AI tools to produce animated videos, often as a way to generate additional income or build a side hustle. Many are creating multiple videos using simple AI tools and paying monthly subscriptions to increase their content production.
However, I also came across several videos where creators discussed how AI-generated content could affect YouTube monetization. Some mentioned that their videos were demonetized or that their channels faced monetization restrictions or rejection.
This made me curious to try the process myself. I decided to create an animated video using one of these AI tools to understand how it actually works.
What I discovered was that several AI video-generation platforms offer monthly subscriptions ranging from approximately $15–25, allowing creators to generate multiple videos depending on their subscription plan and usage limits.
I signed up for a trial and started experimenting. One thing I quickly noticed was the lack of visual consistency and customization. When generating multiple videos, the characters often followed very similar patterns. For example, the baby's facial features, body proportions, and overall animation style looked almost identical across different videos.
It made me realize that many creators might be using similar AI-generated characters and animation styles to produce large volumes of content.
This experiment helped me understand both the convenience and limitations of AI-powered video creation tools.
After experimenting with those tools, I realized that I needed a simple, beginner-friendly tool that would help me create videos based on my own ideas. So, I decided to try creating a video myself first.
As I started exploring the process, I began thinking more about the kind of video experience I actually wanted to create.
What I have noticed about kids' content?
One thing I noticed about kids' content is how easily children get drawn into watching one video after another. Once a video ends, another one starts playing, and this cycle can continue almost endlessly. Sometimes, children find it difficult to stop watching, and when parents turn off the videos, it can lead to frustration or crying because they want to keep watching more.
This made me think about the role of autoplay and continuous video recommendations in kids' content. I wanted to understand how this experience works and explore whether I could build something different: a customized video creation application where I have more control over the content, characters, stories, and overall viewing experience.
So, my first priority was not to create long videos that would keep children watching the screen for extended periods. Instead, I wanted to experiment with a different approach to children's content.
My initial idea:
My initial idea was to start with a short visual introduction, lasting anywhere from 10 seconds to 1 minute, where the characters would introduce the story using facial expressions and gestures. After that, the experience would transition into a full voice-based storytelling format, allowing children to focus more on listening rather than continuously watching the screen.
I wanted to create something similar to a podcast-style experience for kids, where engaging stories, characters, and voiceovers would encourage children to listen and use their imagination instead of spending too much time in front of a screen.
That's where my story began!
Implementing next steps with my Goal:
My ultimate goal is to build a customized application that encourages children to enjoy stories through listening while helping reduce their screen time.
Then I started exploring and experimenting with different options. One of the first things I looked into was how to make illustrated characters lip-sync with the voice, matching their facial movements, expressions, and emotions to the narration.
As I explored further, I realized that I needed an application that could bring together two important capabilities: converting text into voice and generating illustrated images based on my stories or poems for children.
My biggest Challenge:
One of the biggest challenges I faced was finding the right voice options for my videos. I didn't want to use or expose my original voice. Instead, I wanted children to feel as though another child or a friendly cartoon character was telling them the story.
I started thinking about different voice personalities that would make the experience more engaging and relatable for children. For example:
- Young female voices: To represent friendly female cartoon characters.
- Child voices: So children could feel as though another child, similar to their age, was telling the story.
- Adult female voices: To create a warm, gentle, and comforting storytelling experience, similar to listening to a mother or a familiar adult reading a story.
My intention was to create a collection of voice options that would make storytelling feel more natural, friendly, and engaging for children.
That is why having multiple voice options became one of my key requirements when planning my application.
First, I explored several AI voice-generation applications that offered text-to-voice and voice-to-voice conversion. However, most of these platforms required paid subscriptions to access additional features, and their free trials did not give me enough flexibility to experiment with different voice options.
So, I decided to take a different approach and build my own voice-generation application.
Building my own voice-generation application:
When I started building it, I discovered that AI-powered coding tools were very helpful in creating the application's interface, functionality, and interactions. They allowed me to build a good user experience and make changes to the application whenever I wanted.
My first challenge on building my own application: However, I faced another challenge: generating natural, human-like voices. Most of the solutions I initially explored produced robotic-sounding voices or focused mainly on audio processing features such as noise reduction. They didn't provide the kind of natural voice-generation experience I was looking for.
I realized that AI coding tools are excellent for building application interfaces and functionality, but building a voice-generation application requires additional voice technologies and integrations.
Next, I explored AI animation tools. These platforms offered pre-designed characters that could be animated using recorded audio. I could record my own voice, upload it, and the selected character would automatically generate lip movements and facial expressions that matched the audio.
Although this was an interesting approach, it didn't completely solve my problem because I wanted to use different AI-generated voices instead of recording my original voice.
Then, I explored another category of AI tools that could generate better application interfaces, illustrations, and visual assets. These tools helped me explore the visual possibilities, but I was still looking for a way to bring everything together into one customized application.
My biggest learning was that creating an AI-powered video application is not just about building an interface. It requires bringing together multiple capabilities, including voice generation, illustration creation, character animation, and audio synchronization.
I'll share my next experience in the upcoming article.
If you're interested in following my application-building journey and learning how I approach challenges as a UX Researcher and Designer, feel free to follow me!
And if you've already experimented with building similar applications or explored other AI tools, I'd love to hear about your experiences. Please share them in the comments. I'm always happy to learn from others, too!
Originally published by Dev.to WebDev. Aggregated on AIWithGhost for educational purposes — full credit and traffic to the original publisher.



