🛠️ Build your MVP in 7 days
🤝 Find your team on the platform
📈 Hours of educational context
🦾 Generative AI tools to build with
🏅 Further support for winners

AI Startup Hackathon

🛠️ 7 days of Learning and Building
🤝 Come with your team or find one during the Hackathon
🦾 Free access to Generative AI models and Tech Tutorials
🏅 Win a place in the Early Stage StartUp Accelerator program
🚀 Join community of AI experts

Build your AI Startup Hackathon. Episode 2

No-rank

Apprentice

antoni0z

Automated content distribution and generation

Gener8

turit0

A user will provide the link of the podcast he uploaded to youtube. Our project connects to youtube API to be able to extract both the video and the audio. 

Then it converts the audio into text, thanks to whisper and does sentiment analysis for each phrase that is said in the podcast. All this is formatted into a dataframe that is used for more in depth analysis. 

An statistical algorithm calculates the average emotional polarity of each of the clips and select the best three ones. The idea was to optimize for interaction in social media, and emotional polarity guarantees that it will be interesting. 

Finally, it outputs titles, description, subtitle files and already cropped videos to the user, so he can simple copy and paste it in its youtube. The idea for the future would be to automate the content uploads as well. 

This MVP is just a small piece of our vision, and our idea is to build an application that allows massive scale content distribution, identifying the best forms of repurpusing the content a user has an automatically uploading it for them to social media.




Our project allows automatic repurposing of long format podcast into short videos, providing 3 short video fragments, optimized for emotional polarity, so that it maximizes interaction.

Automated AI Content

Automating content generation and distribution powered by AI

Ivanovich

GPT-3 stands for Generative Pre-trained Transformer 3 and it is an autoregressive language model that uses deep learning to produce human-like text. It is the third-generation language prediction model in the GPT-n series created by OpenAI. GPT-3 is currently in open beta.

GPT-3

OpenAI Codex is an artificial intelligence platform that enables developers to translate natural language into code. 

Codex

DALL·E 2 is a new AI system that can create realistic images and art from a description in natural language.

DALL-E-2

One of the best AI labs in the world. By using the Cohere platform we are able to use one of the best and most powerful language models for our own! There are many tutorials, including our own one, that show how to use Cohere.

Cohere

AI21 Labs is a platform for building AI applications. It provides a suite of tools and guides for building AI applications and training models. AI21 Labs helps developers and businesses use cutting-edge language models, with no NLP expertise required. Access powerful language models using an API that makes you smile

AI21 Labs

Whisper is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio and is also a multi-task model that can perform multilingual speech recognition as well as speech translation and language identification.

Whisper

YOLOv7 is the new state-of-the-art object detector in the YOLO family. According to the paper, it is the  most accurate and fastest real-time object detection to date.

YOLOv7

ChatGPT is a large language model trained by OpenAI to generate human-like text in a conversational style. It is a variant of the GPT-3 model, which was specifically designed to be used to generate text in response to user input. The ChatGPT model has been trained on a vast amount of text data, including conversations and other types of human-generated text, which allows it to generate text that is similar in style and content to human conversation. ChatGPT can be used to generate responses to questions, code, make suggestions, or provide information in a conversational manner, and it is able to do so in a way that is often indistinguishable from human-generated text. The initial model has been trained using Reinforcement Learning from Human Feedback (RLHF), using methods similar to InstructGPT, but with slight differences in the data collection setup. The model is trained using supervised fine-tuning, where human AI trainers provided conversations in which they played both sides—the user and an AI assistant. The trainers would have had access to model-written suggestions to help them compose their responses.

Antonio Oliver@antoni0z672

Events attended

Submissions made

About me

I built with

Socials

🤝 Top Collabolators

Arturo Matias Sánchez Horrillo

Ivan Moran

🤓 Submissions

Automated AI Content

👌 Attended Hackathons

AI Startup Hackathon

Build your AI Startup Hackathon. Episode 2

📝 Certificates

AI Startup Hackathon | Certificate