Quality Control for Vinyl Records

The project

Our project aimed to both improve the quality control process of the client, while also making it more efficient, allowing them to scale up production. It works by recording a vinyl record, where after it is automatically analyzed using an algorithm that detects clicks, skipping, and other quality issues. This product allows the client to produce more vinyl records per year with a higher quality, which also means that users will receive fewer faulty vinyl records.

The customer

Our client was Record Industry B.V. We had good communication with the client, since we had biweekly physical meetings and a fixed contact person, who we could reach during office hours by e-mail or by phone

If you have your laptop during a meeting, you don't always have to listen

The team

Our team consisted of six students from different study programmes. Max Gloesener usually lead the client meetings, and was part of the team that developed the quality control algorithm. Ruben Keurentjes was the product owner, and he is also the person that physically visited the client the most. Furthermore, he was part of the backend team that developed the logic behind the program. Foppe van Berkel was our scrum master, and he was part of the quality control algorithm. Dominieq de Koster was the person who took minutes during the client meetings, and he was part of the backend team. Finally, Pepijn Baggen was part of the quality control algorithm team, and Elliot Archer developed the frontend.

As a team, the challenges that we had to overcome were mostly integrating the code of the three different teams. Although we had decided upfront on how communication between the backend, the frontend and the algorithm would take place, there were sometimes issues when integrating all of the code. As a team, we solved this by having good communication between the teams. We are most proud of the product as a whole, since we think that we were able to incorporate all of the client’s wishes, including additional features that were asked for during development.

The technologies

Languages:

• HTML, CSS, JavaScript: Frontend
• Python: Backend

Frameworks and libraries:

• Node.JS: frontend framework for web applications
• React: library for building interactive web pages
• Flask: backend framework for web applications
• Audalign: aligning two sound files
• Soundfile: reading and writing sound files
• Librosa: analysing sound files

Tools:

• Pytest: unit testing of backend modules

Deployment:

• The program can be run on the client’s pc, or possibly an internal server.