no code implementations • 30 Apr 2020 • Alessandro Ilic Mezza, Emanuël. A. P. Habets, Meinard Müller, Augusto Sarti
The performance of machine learning algorithms is known to be negatively affected by possible mismatches between training (source) and test (target) data distributions.
no code implementations • 20 Nov 2018 • Soumitro Chakrabarty, Emanuël. A. P. Habets
In this work, we propose to use systematic dilations of the convolution filters in each of the convolution layers of the previously proposed CNN for expansion of the receptive field of the filters to reduce the computational cost of the method.
no code implementations • 31 Jul 2018 • Soumitro Chakrabarty, Emanuël. A. P. Habets
Supervised learning based methods for source localization, being data driven, can be adapted to different acoustic conditions via training and have been shown to be robust to adverse acoustic environments.
no code implementations • 12 Dec 2017 • Soumitro Chakrabarty, Emanuël. A. P. Habets
The problem of multi-speaker localization is formulated as a multi-class multi-label classification problem, which is solved using a convolutional neural network (CNN) based source localization method.
1 code implementation • 12 Dec 2017 • Fabian-Robert Stöter, Soumitro Chakrabarty, Bernd Edler, Emanuël. A. P. Habets
The task of estimating the maximum number of concurrent speakers from single channel mixtures is important for various audio-based applications, such as blind source separation, speaker diarisation, audio surveillance or auditory scene classification.
Audio and Speech Processing Sound
1 code implementation • 2 May 2017 • Soumitro Chakrabarty, Emanuël. A. P. Habets
Since only the phase component of the input is used, the CNN can be trained with synthesized noise signals, thereby making the preparation of the training data set easier compared to using speech signals.