Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur Graph Search.
In this project, I developed a browser based interface for prototyping real-time processing algorithms using embedded multi channel audio acquisition platforms. The interface allows to interact in an easy manner with a GNU/Linux enabled single board computer and to run algorithms in real-time on distant audio streams from a host computer. It creates a bridge between the user’s computer and the audio acquisition board. More precisely, the interface let us to: - change the board configuration; - write Python code; - execute it in real-time on the audio streams; - visualize data (typically outputs of the algorithm). The project is composed of three main components, communicating together using WebSockets: - daemons in C running on the board: they transmit the audio streams and listen for configuration changes; - a Python daemon: it executes Python code sent by the interface; - the interface itself. Finally, a Python module helps the user to access the audio streams, the configuration, and to create data visualizations. This module connects to the C daemons (to receive the streams and the configuration) and to the browser to send data in real-time for visualizations. The technical documentation is included in the Appendix of this report. It is also available online: https://lcav.github.io/easy-dsp/, as well as the code repository: https: //github.com/LCAV/easy-dsp.