Image and Audio Preprocessing with LLMs
In this chapter, we delve into the preprocessing of unstructured data, specifically focusing on images and audio. We explore various techniques and models designed to extract meaningful information from these types of media. The discussion includes a detailed examination of image preprocessing methods, the use of optical character recognition (OCR) for extracting text from images, the capabilities of the BLIP model for generating image captions, and the application of the Whisper model for converting audio into text.
In this chapter, we’ll cover the following topics:
- The current era of image preprocessing
- Extracting text from images
- Handling audio data
