yigitkonur/swift-ocr-llm-powered-pdf-to-markdown

Stars
24
Rank 986,245 (Top 20 %)
Language
Python
Created 2 months ago
Updated 2 months ago

yigitkonur/swift-ocr-llm-powered-pdf-to-markdown

yigitkonur

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

An open-source OCR API that leverages OpenAI's powerful language models with optimized performance techniques like parallel processing and batching to deliver high-quality text extraction from complex PDF documents. Ideal for businesses seeking efficient document digitization and data extraction solutions.

telnyx-whisper-bulk-caller

A script that uses Telnyx API to make bulk calls to a given list of numbers and analyzes the audio recordings with Whisper. This way, we don’t have to read all the transcriptions.

write-into-menubar

Write whatever you want into OSX menubar by using BitBar + Alfred Powerpack.

context-aware-srt-translation-gpt

A repository trying to translate subtitles with GPT 3.5 Turbo without losing context (using the dynamic window context method).

data-preparation-for-fine-tuning

A Python project for preparing and analyzing datasets from JSONL files. It includes tools for shuffling, categorizing, and generating reports on dataset content.

rust-hyper-load-balanced-api-client

A high-performance Rust tool for sending API requests (to LLMs in my case) with built-in weighted load balancing, retry mechanisms, and rate limiting. Using hyper for fast request handling, it manages large volumes of asynchronous requests and is optimized for 10K request per second.

fastapi-http-proxy-with-caching

A FastAPI-based HTTP proxy with request caching using Redis, designed to forward requests while caching responses for efficient repeated queries.

bulk-openai-embeddings-creator

A multi-threaded script & CLI tool for generating embeddings from text using multiple OpenAI endpoints. Supports resuming from previously processed data and customizable thread configurations

cluster-by-similarity-high-dim-vectors

Python script for automated clustering of embeddings with DBSCAN, using cosine similarity for flexible, size-agnostic grouping (I hate K-means); outputs analysis metrics.

copilot-pr-first-comment-updater

Python script that appends the copilot:all label to the first comments of closed pull requests in a specified GitHub repository, allowing easy tracking and management of processed PRs.

n8n-docker-ffmpeg

Repository for setting up n8n with ffmpeg using Docker Compose, including beginner-friendly instructions and systemd service configuration for automatic startup.

pineconedb-appscript-integration-for-sheets

A Google Apps Script custom function to fetch similar categories from a vector database using OpenAI and Pinecone APIs. This function can be used directly in Google Sheets.

go-native-squid-proxy

GoNativeSquidProxy is a high-performance, scalable proxy server fully written in Go, designed to efficiently handle HTTP/HTTPS requests as a modern alternative to Squid.

Go-JSON-AzureSearch-Prepper

A Go utility for processing and combining JSON files, making them ready for integration with Azure Search AI.

kv-backup-enhanced

A script to help you to download 1200 Cloudflare KV records per 5 minute (restricted due to Cloudflare's global rate limit)

GoogleSheets-Translator

Automate translations in Google Sheets using Google Translate, optimized for handling large datasets with efficient batch processing (by using =GOOGLETRANSLATE formula to handle large amount of data effectively)

translation-cli-by-openai-api

This CLI tool makes it easy to translate a large number of strings from an XLSX file into many languages using OpenAI. It has powerful features like a progress bar and the ability to customize the file structure.