AI Talebook

Inspiration

As a kid, I enjoyed listening to my mother reading me stories and fairy tales. I always wondered if I could hear her voice, listed to the story when she was not there.

What it does

AI TaleBook is a complete story experience. It has AI generated:

• Storyline, a plot with a positive moral
• Voiceover
• Background Music
• Relevant Scenes as images
• A complete Video with these components
• A Speech to Text Question Answer function that fulfils a child's curiosity

How we built it

We used OpenAI model for Story generation, Mubert for background music, UberDuck for the voiceover, Replicate for image generation, MoviePy for generating movie , Assembly AI Speech to Text and OpenAI for question answering.

Challenges we ran into

We faced following challanges:

• Mainly Limited Computation and resources.
• We adopted Moviepy over FILM.
• We used Prompt to Image instead of Prompt to Video or Image to Image models

Accomplishments that we're proud of

At the end we completed the pipeline. Showed it to our family and they were amazed to see the capability of AI.

What we learned

While doing this project we understood computation limitations that we faced, we adopted new solutions discarding what we original planned to use. We learned that AI is very good incase of generation (text, image as of yet). Furthermore continuous research and study is a must to keep up with this every-day-new-model age of AI.

What's next for AI Talebook

Well we have a complete story experience here. Each and every components requires improvement. With advances in AI and technology, the AI TaleBook will surely prove its worth.

Built With

• assemblyai
• python
• pytorch
• replicate