Back to portfolio

AI & Automation ยท 2022

AI Music Credits Extractor

An AI-powered tool for extracting music credits from user-generated content.

The AI Music Credits Extractor is a tool developed for Rockit.it to transform unstructured music credits submitted by users into structured data that can be used to build a database of music professionals.

On Rockit, artists upload their albums and typically include credits in free-text form: producers, musicians, recording studios, mastering engineers and other contributors. Because these credits are written in many different styles and formats, they are difficult to process automatically with traditional parsing techniques.

This project uses a large language model via the OpenAI API to interpret the text and extract meaningful entities such as names, roles and relationships. The extracted information is then normalized and converted into structured records that populate a database of people involved in the production of the album.

The goal of the system is to progressively build a richer graph of collaborations within the independent music scene, linking artists with producers, technicians and other contributors across different releases.

The tool is implemented in PHP and integrated into the existing Rockit infrastructure, where it processes user-generated content and assists in maintaining consistent credit information across the platform.

This is an internal tool, not not publicly availble.

Technologies

  • MySql
  • Openai API
  • PHP