Home Page

Description

This project aims to streamline the creation process for creators, allowing many to create a whole product with just a drop of inspiration. With our website, users can create a full-fledged video with just their ideas expressed in textual form, providing a safe haven for non-artistically inclined creators (includes skills like directing, video editing, etc).

For example, with just a prompt and the click of a button, users can generate a thumbnail (image), content (video), a narrator (audio) and background audio all at once to post on their desired platform. Furthermore, we allow easy customisation and provide choices, so users are not stuck with only one option that does not fit their requirements.

Problem Statement

In the scenarios of creating and consuming streaming media content, generative Al technologies can be utilized for content optimization, information extraction, and style transformation, to refine content across various media platforms. With these technologies, we can cater to the preferences of diverse audiences, as well as facilitate creators in producing higher quality content more efficiently.

Features and Functionality

Content Generation

Image generation with prompt (Optional parameter: negative prompt)
Video generation with prompt
Speech generation with prompt (Optional parameters: language, gender of speaker, voice source)
Sound Effect Generation with prompt

Additional functionality

Change generated content if not satisfied with the current one
- Sound effect can only cycle between 2 options
- Speech has limited options so it will switch back to the original one eventually
Combine generated content into a final video
- Can input own media files to combine as well

Tools used

GitHub
VS Code
Bootstrap v5.0.2
HuggingFace

APIs used

HuggingFace Spaces gradio API
1. Aqcua/TextToImage-AISDXLTURBO
2. stabilityai/stable-diffusion-3-medium
3. tonyassi/voice-clone
4. k2-fsa/text-to-speech
5. BestWishYsh/MagicTime
6. https://videocrafter-videocrafter.hf.space/
7. https://haoheliu-audioldm2-text2audio-text2music.hf.space/
8. eagle0504/stable-audio-demo