Turn image to text. Write a story based on that text. Read the story
- Load local .jpg image
- Generate text description of the image with Salesforce/blip-image-captioning-large model based HuggingFace pipeline
- Generate more text (a story) based on the image description using GPT2 model based HuggingFace pipeline
- Convert story to .flac audio file with HuggingFace Inference API for kan-bayashi_ljspeech_vits model