Now showing items 1-20 of 29

    • A multi-microphone approach to speech processing in a smart-room environment 

      Abad Gareta, Alberto (Universitat Politècnica de Catalunya, 2007-06-29)
      Doctoral thesis
      Open Access
      Els avenços recents en tecnologia informàtica i processament de la parla i del llenguatge, entre altres, han fet possible que noves maneres de comunicació entre les persones i les màquines comencin a semblar factibles. ...
    • A system for robust iris recognition 

      Pérez Parera, Dídac (Universitat Politècnica de Catalunya, 2012-02-13)
      Master thesis (pre-Bologna period)
      Restricted access - author's decision
      [ANGLÈS] Biometric technologies are becoming important due to the increasing demand for flexible and robust systems for identification and verification of individuals. In recent years, some biometrics has been proposed ...
    • Caracterització de locutors usant deep learning 

      Alcón Doganoc, Miguel (Universitat Politècnica de Catalunya, 2018-06-26)
      Bachelor thesis
      Open Access
      Desenvolupament d'un sistema computacional capaç de comparar dos fitxers d'àudio, on parla una persona, i determinar si comparteixen locutor, amb un error òptim del 29,77%; i d'una interfície gràfica que l'utilitza.
    • Deep learning bottle-neck features for speaker recognition 

      Cumalat Puig, Eudald (Universitat Politècnica de Catalunya, 2018-10)
      Bachelor thesis
      Open Access
      Speaker recognition is a very useful and powerful technology that has many interesting security applications, which makes it an investigation field to pour on many efforts. Recognizing whether a person is who he/she claims ...
    • Deep learning for i-vector speaker and language recognition 

      Ghahabi Esfahani, Omid (Universitat Politècnica de Catalunya, 2018-05-29)
      Doctoral thesis
      Open Access
      Over the last few years, i-vectors have been the state-of-the-art technique in speaker and language recognition. Recent advances in Deep Learning (DL) technology have improved the quality of i-vectors but the DL techniques ...
    • Deep Neural Networks for Channel Compensated i-Vectors in Speaker Recognition 

      Jiménez Sanfiz, Albert (Universitat Politècnica de Catalunya, 2014-06)
      Bachelor thesis
      Open Access
      This thesis explores the application of channel-compensation techniques in speaker verification and the posterior combination with deep learning technologies. The idea is to reduce the degradation of the performance due ...
    • Detection and handling of overlapping speech for speaker diarization 

      Zelenák, Martin (Universitat Politècnica de Catalunya, 2012-01-31)
      Doctoral thesis
      Open Access
      For the last several years, speaker diarization has been attracting substantial research attention as one of the spoken language technologies applied for the improvement, or enrichment, of recording transcriptions. ...
    • Discriminative features for GMM and i-vector based speaker diarization 

      Zewoudie, Abraham Woubie (Universitat Politècnica de Catalunya, 2017-09-20)
      Doctoral thesis
      Open Access
      Speaker diarization has received several research attentions over the last decade. Among the different domains of speaker diarization, diarization in meeting domain is the most challenging one. It usually contains spontaneous ...
    • Emociones en señales de voz: reconocimiento con redes neuronales profundas. 

      Hernández Leal, Victor Emilio (Universitat Politècnica de Catalunya, 2021-10-27)
      Bachelor thesis
      Open Access
      In recent years the research effort of different tasks is through neural network techniques. This work continues along this line and explores its possibilities in the task of speech emotion recognition (SER). In this work, ...
    • Feature extraction for speaker diarization 

      Negre Rabassa, Enric (Universitat Politècnica de Catalunya, 2016-04-15)
      Master thesis (pre-Bologna period)
      120 months embargo
      Feature extraction for speaker diarization using different databases
    • Fusing prosodic and acoustic information for speaker recognition 

      Farrús Cabeceran, Mireia (Universitat Politècnica de Catalunya, 2008-10-29)
      Doctoral thesis
      Open Access
      El reconeixement automàtic del locutor és la utilització d’una màquina per identificar un individu a partir de d’un missatge parlat. Recentment, aquesta tecnologia ha experimentat un increment en l’ús de diverses aplicacions ...
    • Identificació de veu mitjançant xarxes neuronals profundes implementades sobre FPGA 

      Palet Gual, Marc (Universitat Politècnica de Catalunya, 2016-04-29)
      Bachelor thesis
      Open Access
      El projecte presenta un sistema d'identificació del gènere de l'interlocutor a partir de la veu basat en una xarxa neuronal profunda implementada en un dispositiu FPGA. L'objecte d'estudi és el rendiment la latència i la ...
    • Integración de biometría de voz en un sistema de pago por teléfono: verificación del locutor dependiente del texto 

      Masana de Bouffard, Judit (Universitat Politècnica de Catalunya, 2016-09)
      Bachelor thesis
      Open Access
      The aim of this project is obtain a text-dependent speaker verification system using speech biometrics. We recover software created by the Department of Signal and Communications Theory, which had been modified. It has ...
    • Integración de tecnologías de audio en entornos inteligentes 

      Lendoiro Rodríguez, Diego (Universitat Politècnica de Catalunya, 2013-12-16)
      Master thesis (pre-Bologna period)
      Open Access
      Anglès: Audio technologies are a key part in the development of smart environments. As for the human beings, audio tecnologies, are a great information resource from which big amounts of data can be obtained. This data ...
    • Integration of speech biometrics in a phone payment system: text-independent speaker verification 

      Barón Garcia, Anna (Universitat Politècnica de Catalunya, 2016-09)
      Bachelor thesis
      Open Access
      Nowadays, the integration of biometrics in security systems is a prominent research and application field. Also, it is clear that speech is the most common form of communication, which makes a swell candidate. While using ...
    • Normalización estadística para fusión biométrica multimodal 

      Ejarque Monserrate, Pascual (Universitat Politècnica de Catalunya, 2011-03-17)
      Doctoral thesis
      Open Access
      Los sistemas de reconocimiento biométrico utilizan ciertas características humanas como la voz, los rasgos faciales, la huella dactilar, el iris o la geometría de la mano para identificar a un individuo o verificar su ...
    • Noves característiques de veu per la diarització de parlants 

      Casanovas Duch, Artemi (Universitat Politècnica de Catalunya, 2015-10)
      Bachelor thesis
      Open Access
      Voice features fusion has been successfully used to both identify and verify speakers and to detect voice pathologies. Nowadays, it is also being tested in speaker diarization task. The following thesis shows a study on ...
    • Online mail classifier using Deep Learning 

      Gil Garcia, Enric (Universitat Politècnica de Catalunya, 2022-02-03)
      Master thesis
      60 months embargo
      This thesis has been developed at NTTData for an external client. The client requires an autonomous mail classifier since they currently classify mails manually. The thesis consists in the development of a Deep Learning ...
    • Parallel scalability of face detection in heterogeneous multithreaded architectures 

      Oro García, David (Universitat Politècnica de Catalunya, 2020-11-17)
      Doctoral thesis
      Open Access
      Recently, facial recognition systems have become extremely popular and deployments of this technology are now ubiquitous. Applications ranging from access control to automated surveillance of video feeds rely on facial ...
    • Predicting emotion in speech: a Deep Learning approach using Attention mechanisms 

      Aromí Leaverton, Daniel (Universitat Politècnica de Catalunya, 2021-06)
      Bachelor thesis
      Open Access
      Speech Emotion Recognition (SER) has recently become a popular field of research because of its implications in human-computer interaction. In this study, the emotional state of the speaker is successfully predicted by ...