<Session_Outline/>
At Soundtrack Your Brand we face the challenge of having to build a state-of-the-art music recommendation system, whilst not having much interactive usage data to work with. So what does it mean to build machine-learning models that only have audio to learn from? And what are the characteristics that set music apart from other audio data like speech?
<Key_Takeaways/>
- Music has additional characteristics compared to other audio that can be utilised
- There is a lot of interesting overlap between the physics of sound and music theory
- Several layers of embeddings can limit the need for high-quality labeled data
————————————————————————————————————————————————————
<Speaker_Bio/>
Tobias Duin – Senior Machine Learning Engineer | Soundtrack Your Brand
Tobias has been in the Data Science and AI industry for 7 years, where he has worked with a variety of topics like real-time bidding and financial forecasting. His role ranged from product manager to working with the nitty gritty details of the algorithms. Right now he is combining deep-learning with his passion for music at Soundtrack Your Brand.
Day 2 | 19th of May – Engineering
Tobias Duin – Senior Machine Learning Engineer | Soundtrack Your Brand