Please note! Course description is confirmed for two academic years, which means that in general, e.g. Learning outcomes, assessment methods and key content stays unchanged. However, via course syllabus, it is possible to specify or change the course execution in each realization of the course, such as how the contact sessions are organized, assessment methods weighted or materials used.
Students will learn recent deep learning & artificial intelligence (AI) models and network architectures for audio, work on course exercises and start building AI applications for their own purposes. Students will also learn and practice preparing data sets and traning deep learning models using cluster network in Aalto University. Further, students will gain an understanding of the differences in input, computational cost and sonic characteristics between the different models, which will help formulate a course project.
Schedule: 20.04.2021 - 07.05.2021
Teacher in charge (valid 01.08.2020-31.07.2022): Koray Tahiroglu
Teacher in charge (applies in this implementation): Koray Tahiroglu
Contact information for the course (applies in this implementation):
CEFR level (applies in this implementation):
Language of instruction and studies (valid 01.08.2020-31.07.2022):
Teaching language: English
Languages of study attainment: English
CONTENT, ASSESSMENT AND WORKLOAD
In Deep Learning with Audio, we will introduce students to the state of the art in deep learning models for sound and music generation, giving an overview of recent artificial intelligence (AI) implementations such as WaveNet, WaveGAN, Google Magenta (NSynth, GANSynth, Melody RNN...) and audio style transfer. There will be hands-on exercises on each course topic. We will provide code templates that integrate the functionality from open source deep learning audio projects, such as Google's Magenta, into Pure Data programming environment. We will also provide detailed setup instructions and automated scripts to make installation of the required tools as easy as possible (for Pure Data, Python, Conda, Magenta, PyExt). Students will further explore a particular model and incorporate it into their own project work.
Assessment Methods and Criteria
The course consists of lectures, exercises, reading materials, tutoring individual or group works. Students will submit their documented project work and ~ 750 words learning diary, both grounds the course examination and final grade. Each student project work will be assessed with the following criteria: Design Values, Aesthetics and Originality; UI design and Production Values; Code Design Quality; Project Analysis - Depth of Understanding; Idea generation and implementation; and Presentation style.
This course is a project-based course. In addition to 36h of contact teaching, at the end of the course, students will submit and present their projects.
We strongly suggest students to take DOM-E5074 Composing with Data Flow Programming and DOM-E5129 Intelligent Computational Media courses before they register to Deep Learning with Audio course. DOM-E5074 and DOM-E5129 will provide the basic knowledge of the tools and methods that will be used in this course.
- Teacher: Koray Tahiroglu