Fuzzy matching and more functionality for spaCy.
современный голосовой ассистент для Windows
High-performance fuzzy string matching for .NET with bit-parallel Levenshtein/Indel algorithms, FuzzyWuzzy-compatible scorers, and fast extraction APIs.
DuckDB extension for fuzzy string matching, edit distances, search, and record linkage using RapidFuzz.
Fast Batch String Matching in Python (Levenshtein, Jaro-Winkler, Hamming) with Zero Cache Misses - made for Python, written in C++
✅ completed | Voices assistant for windows managing system applications
Offline (EM/F1/BLEU/ROUGE/perplexity/semantic-similarity) and online (FastAPI + llama.cpp + MySQL cache + Streamlit dashboard) evaluation metrics for LLM applications.
Fast Scalable Dedupe - Fuzzy Matching With Opensearch + nmslib + Rapidfuzz
Match address data files together (CSV, XLSX, Parquet) using fuzzy matching (rapidfuzz) and an (optional) graphical user interface. Built for UK addresses. Best used in combination with the uk_address_matcher repo: https://github.com/moj-analytical-services/uk_address_matcher
Automate job application tracking by parsing Gmail emails and logging them to Google Sheets by using Gmail API, Google Sheets API, OAuth 2.0, Gemini LLM and Python.
NovelNudge is a book recommendation engine that embeds titles, descriptions, authors, and genres using SentenceTransformers. It combines these vectors and ranks similar books with cosine similarity and fuzzy title matching.
A simple and efficient spelling correction system that uses Python's rapidfuzz library to find and correct misspelled sentences by matching them with the closest correct ones from a given dataset.
Fast fuzzy string matching for Node.js and the browser
AI-powered TTB label verification tool for compliance agents. Upload label images (batch supported), automatically compare brand name, ABV, government warning, and 4 more fields against application data. Color-coded checklist with match/review/mismatch verdicts, confidence scores, and processing time. Built with Streamlit + EasyOCR.
An agentic music rights intelligence platform for catalog search, rights management, compliance, and version tracking.
Config-driven student-records data-integrity & governance pipeline: bulk-ingest, validate, de-duplicate (exact + fuzzy), audit, and report on synthetic student data — the operational, governance side of data work.
Self-hosted NotebookLM alternative for Notion: Telegram bot answers questions over your workspace with citations.
To associate your repository with the rapidfuzz topic, visit your repo's landing page and select "manage topics."