no code implementations • 14 Jun 2023 • David Diaz-Guerra, Archontis Politis, Antonio Miguel, Jose R. Beltran, Tuomas Virtanen
Conventional recurrent neural networks (RNNs), such as the long short-term memories (LSTMs) or the gated recurrent units (GRUs), take a vector as their input and use another vector to store their state.
1 code implementation • 24 Mar 2023 • Carlos Hernandez-Olivan, Sonia Rubio Llamas, Jose R. Beltran
In the past, there have been several works that attempt to segment music into the audio and symbolic domains, however, the identification and segmentation of the music structure at different levels is still an open research problem in this area.
no code implementations • 25 Oct 2022 • Carlos Hernandez-Olivan, Javier Hernandez-Olivan, Jose R. Beltran
How humans perceive and understand music is still being studied and is crucial to develop artificial intelligence models that imitate such processes.
2 code implementations • 31 Mar 2022 • David Diaz-Guerra, Antonio Miguel, Jose R. Beltran
In this paper, we present a new model for Direction of Arrival (DOA) estimation of sound sources based on an Icosahedral Convolutional Neural Network (CNN) applied over SRP-PHAT power maps computed from the signals received by a microphone array.
no code implementations • 28 Mar 2022 • Carlos Hernandez-Olivan, Jorge Abadias Puyuelo, Jose R. Beltran
We use this method to compare state-of-the-art models for music composition with deep learning.
1 code implementation • 27 Aug 2021 • Carlos Hernandez-Olivan, Jose R. Beltran
Generating a complex work of art such as a musical composition requires exhibiting true creativity that depends on a variety of factors that are related to the hierarchy of musical language.
1 code implementation • 13 Jul 2021 • Carlos Hernandez-Olivan, Jose R. Beltran
It has been possible to assess the ability to classify instruments by timbre even if the instruments are playing the same note with the same intensity.
2 code implementations • 17 Aug 2020 • Carlos Hernandez-Olivan, Jose R. Beltran, David Diaz-Guerra
The objective of this work is to establish a general method of pre-processing these inputs by comparing the inputs calculated from different pooling strategies, distance metrics and audio characteristics, also taking into account the computing time to obtain them.
2 code implementations • 16 Jun 2020 • David Diaz-Guerra, Antonio Miguel, Jose R. Beltran
In this paper, we present a new single sound source DOA estimation and tracking system based on the well-known SRP-PHAT algorithm and a three-dimensional Convolutional Neural Network.
3 code implementations • 26 Oct 2018 • David Diaz-Guerra, Antonio Miguel, Jose R. Beltran
The Image Source Method (ISM) is one of the most employed techniques to calculate acoustic Room Impulse Responses (RIRs), however, its computational complexity grows fast with the reverberation time of the room and its computation time can be prohibitive for some applications where a huge number of RIRs are needed.