Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

An Open-Set Recognition and Few-Shot Learning Dataset for Audio Event Classification in Domestic Environments

  • Naranjo-Alcaraz, Javier
  • Perez-Castanos, Sergi
  • Zuccarello, Pedro
  • Torres, Ana M.
  • Ferri, Francesc J.
  • Cobos, Maximo
  • López Monfort, José Javier|||0000-0001-6884-5577
[EN] The problem of training with a small set of positive samples is known as few-shot learning (FSL). It is widely known that traditional deep learning algorithms usually show very good performance when trained with large datasets. However, in many applications, it is not possible to obtain such a high num- ber of samples. This paper deals with the application of FSL to the detection of specific and intentional acoustic events given by different types of sound alarms, such as door bells or fire alarms, using a limited number of samples. These sounds typically occur in domestic environments where many events corre- sponding to a wide variety of sound classes take place. Therefore, the detection of such alarms in a practi- cal scenario can be considered an open-set recognition (OSR) problem. To address the lack of a dedicated public dataset for audio FSL, researchers usually make modifications on other available datasets. This pa- per is aimed at providing the audio recognition community with a carefully annotated dataset1 for FSL in an OSR context comprised of 1360 clips from 34 classes divided into pattern sounds and unwanted sounds. To facilitate and promote research on this area, results with state-of-the-art baseline systems based on transfer learning are also presented., This work was supported by the EU Horizon 2020 programme [grant No 779158] . Grants DIN2018-009982, PTQ-17-09106, RTI2018-097045-B-C21/C22 funded by MCIN/AEI/10.13039/50110 0 011033, the latter also by "ERDF A way of making Europe ". Grants TED2021-131003B-C21/C22 funded by MCIN/AEI/10.13039/50110 0 011033 and by the "EU Union NextGenerationEU/PRTR ". Grants AICO/2020/154 and AEST/2020/012, funded by GVA. The authors acknowledge also the Artemisa computer resources funded by the EU ERDF and Comunitat Valenciana, and the technical support of IFIC (CSIC-UV) . Authors J. Naranjo, S. Perez and P. Zuccarello were working at Visualfy when this work was done, but they are now with the Instituto Tecnologico de Informatica (ITI) , Tyris AI and ITI, respectively.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

An Efficient Implementation of Parallel Parametric HRTF Models for Binaural Sound Synthesis in Mobile Multimedia

  • Belloch, José A.
  • Badia, Jose M.
  • Cobos, Máximo
  • Ramos Peinado, Germán|||0000-0001-7152-0044
[EN] The extended use of mobile multimedia devices in applications like gaming, 3D video and audio reproduction, immersive teleconferencing, or virtual and augmented reality, is demanding efficient algorithms and methodologies. All these applications require real-time spatial audio engines with the capability of dealing with intensive signal processing operations while facing a number of constraints related to computational cost, latency and energy consumption. Most mobile multimedia devices include a Graphics Processing Unit (GPU) that is primarily used to accelerate video processing tasks, providing high computational capabilities due to its inherent parallel architecture. This paper describes a scalable parallel implementation of a real-time binaural audio engine for GPU-equipped mobile devices. The engine is based on a set of head-related transfer functions (HRTFs) modelled with a parametric parallel structure, allowing efficient synthesis and interpolation while reducing the size required for HRTF data storage. Several strategies to optimize the GPU implementation are evaluated over a well-known kind of processor present in a wide range of mobile devices. In this context, we analyze both the energy consumption and real-time capabilities of the system by exploring different GPU and CPU configuration alternatives. Moreover, the implementation has been conducted using the OpenCL framework, guarantying the portability of the code., This work was supported in part by the Spanish Ministry of Economy and Competitiveness under Grant RTI2018-097045-B-C21, Grant RTI2018-097045-B-C22, Grant TIN2017-82972-R, and Grant ESP2015-68245-C4-1-P, and in part by the Universitat Jaume I Project UJI-B2019-36.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Compensating first reflections in non-anechoic head-related transfer function measurements

  • López Monfort, José Javier|||0000-0001-6884-5577
  • Gutiérrez-Parera, Pablo|||0000-0002-5349-5698
  • Cobos, Maximo
[EN] Personalized Head-Related Transfer Functions (HRTFs) are needed as part of the binaural sound individ- ualization process in order to provide a high-quality immersive experience for a specific user. Signal processing methods for performing HRTF measurements in non-anechoic conditions are of high interest to avoid the complex and inconvenient access to anechoic facilities. Non-anechoic HRTF measurements capture the effect of room reflections, which should be correctly identified and eliminated to obtain HRTFs estimates comparable to ones acquired in an anechoic setup. This paper proposes a sub-band frequency-dependent processing method for reflection suppression in non-anechoic HRTF signals. Array processing techniques based on Plane Wave Decomposition (PWD) are adopted as an essential part of the solution for low frequency ranges, whereas the higher frequencies are easily handled by means of time-crop windowing methods. The formulation of the model, extraction of parameters and evaluation of the method are described in detail. In addition, a validation case study is presented showing the suppression of reflections from an HRTF measured in a real system. The results confirm that the method allows to obtain processed HRTFs comparable to those acquired in anechoic conditions., This work has received funding from the Spanish Ministry of Science, Innovation and Universities, through projects RTI2018097045-B-C21 and RTI2018-097045-B-C22, and Generalitat Valenciana under the AICO/2020/154 project grant.