The term “differentiable digital signal processing” describes a family of techniques in which
loss function gradients are backpropagated through digital signal processors, facilitating …

I am delighted to be in a position to write this preface: it is the last task that I need to get done
before I submit the manuscript for production. I have just gone through the checklist. All good …

La séparation de sources sonores consiste à extraire une ou plusieurs sources présentant
un attrait significatif d'un enregistrement contenant plusieurs sources sonores. Ces …

In the recent past years, deep learning based machine learning systems have demonstrated
remarkable success for a wide range of learning tasks in multiple domains such as computer …

Abstract Robotic Musicianship research at Georgia Tech Center for Music Technology
(GTCMT) focuses on the construction of autonomous and wearable robots that analyze …

Estimating motion and surrounding geometry of a moving camera remains a challenging
inference problem. From an information theoretic point of view, estimates should get better …

We present UTACO, a singing synthesis model based on an attention-based sequence-to-
sequence mechanism and a vocoder based on dilated causal convolutions. These two …

Music-generative AI raises multiple challenges particularly related to the work of artists, the
existing music industry model, the role of AI in creative processes, and the discussion of …

We introduce a data-driven approach to automatic pitch correction of solo singing
performances. The proposed approach predicts note-wise pitch shifts from the relationship …

Informed source separation has recently gained renewed interest with the introduction of
neural networks and the availability of large multitrack datasets containing both the mixture …