knowledgegpt is an AI-powered tool that extracts information from websites, PDFs, PowerPoint files, and documents, as well as audio and video sources using speech-to-text technology. It creates prompts that can be utilized by OpenAIs GPT-3 model to generate answers that are stored in a database for future reference. To use knowledgegpt, you can install it from PyPI using `pip install knowledgegpt` or directly from the repository. The tool also requires downloading en_core_web_sm language model for parsing. Once installed, you can import the library and use different extractors for different sources of information such as WebScrapeExtractor, PDFExtractor, PowerpointExtractor, DocsExtractor, and YoutubeAudioExtractor. You can prompt the OpenAI model with a query and receive an answer along with the prompt and additional messages. knowledgegpt also supports setting API keys and parameters for extraction and saving results in a MongoDB database.
