Audio Processing Using Python Laboratory
General Information
Tutors: Christoph Weyer
Date: 10 practice sessions á 4 hours
(further details are announced at the start of the semester)
Requirements: Completed Bachelor Degree
Manuscript: All needed documents are provided free of charge
Language: English and German
Registration: via RWTHonline
Dates
Introductory session (Mandatory):
Thursday, 11 April 2024
02.00 PM - 03.30 PM
The lab takes plan in person. The total number of participants is limited. Who will be assigned a spot on the lab is decided in the (mandatory!) introductory session. The introductory session will take place virtually on the 11 April 2024 between 2:00 and 3:30 PM via zoom. The invitation link will be provided in advance to all students registered in RWTH-online.
Lab time takes place on Thursdays from 2:00-6:00 PM
Lab 1: 18.04.2024
Lab 2: 25.04.2024
Lab 3: 02.05.2024
Lab 4: 16.05.2024
Lab 5: 06.06.2024
Lab 6: 13.06.2024
Lab 7: 20.06.2024
Lab 8: 27.06.2024
Lab 9: 04.07.2024
Lab 10: 11.07.2024
Content
The programming language Python enables rapid and comprehensive development of prototypes for signal processing and machine learning. Powerful libraries are available since a few years. Because of the scope of possibilities and the open-source license, Python has found widespread use in research groups and departments in academia and industry. This module allows students to experience Python in order to prepare for working in science or industry.
This lab addresses two core targets: learning programming techniques in Python as well as applying fundamental techniques in signal processing and machine learning. Both targets are pursued in parallel in the context of audio signal processing; the Python libraries and the machine learning methods are oriented towards applications from this field. The lab is organized in a series of prepared experiments on diverse use cases, for instance:
- Signal analysis
- Filter design
- Adaptive filters and noise reduction
- Multi-channel and spatial audio processing
- Pattern recognition using machine learning
- Classification and regression techniques