This is a YouTube Comment Bot that utilizes OpenAI's GPT-4 model to generate and post comments on YouTube videos based on a user's opinion. The bot can also perform searches and extract video comments.
- googleapiclient
- google-auth-oauthlib
- googleapiclient.errors
- openai
- pytube
- whisper
- transformers
- PIL (Pillow)
This file contains the YouTubeSearch class, which is responsible for performing YouTube video searches, extracting video comments, and posting generated comments to YouTube. It also handles creating and updating a JSON file containing video search results and transcriptions.
This file contains the AI class, which interacts with the OpenAI GPT-4 model to generate comments or search queries based on the given prompts and context.
This file contains utility functions such as load_prompts() and parse_bullets() that assist in reading data from text files and parsing information.
This file contains the transcribe_video() function that downloads and transcribes the audio from a YouTube video using the Whisper ASR model.
This file contains the generate_caption() function, which utilizes the Hugging Face Transformers library to generate captions for images using the VIT-GPT2 model.
This file contains the get_code_from_file() function that reads and returns the content of a Python file. It also creates a 'code.txt' file containing the codebase in a readable format.
This file is the entry point of the application, where the user provides their opinion and the bot starts interacting with YouTube videos.
- Install the required dependencies.
- Set up the API keys for OpenAI and YouTube in the
main.pyfile. - Run
main.pyand input your opinion when prompted. - The bot will perform searches, generate comments, and interact with YouTube videos based on the given opinion.
Note: This codebase is meant for educational purposes and should not be used to manipulate or influence opinions on social media platforms.