Now showing items 21-40 of 133

    • Comportamiento de la transformacion bilineal de frecuencias en reconocimiento de habla ruidosa 

      Hernando Pericás, Francisco Javier; Nadeu Camprubí, Climent; Riu, D. (. AERFAI, 1992)
      Conference report
      Open Access
    • Comportamiento de la transformación bilineal de frecuencias en reconocimiento de habla ruidosa 

      Hernando Pericás, Francisco Javier; Nadeu Camprubí, Climent (1992)
      Conference report
      Open Access
    • Corpus selection 

      Adda, Gilles; Barras, Claude; Kernal Ekenel, Hazim; Morros Rubió, Josep Ramon; Hernando Pericás, Francisco Javier (2013-03-31)
      Research report
      Open Access
      Entregable del proyecto Collaborative Annotation of multi-MOdal, MultI-Lingual and multi-mEdia documents. This document describes the different corpora that will be used during the Camomile project
    • Deep belief networks for i-vector based speaker recognition 

      Ghahabi Esfahani, Omid; Hernando Pericás, Francisco Javier (Institute of Electrical and Electronics Engineers (IEEE), 2014)
      Conference report
      Restricted access - publisher's policy
      The use of Deep Belief Networks (DBNs) is proposed in this paper to model discriminatively target and impostor i-vectors in a speaker verification task. The authors propose to adapt the network parameters of each speaker ...
    • Deep learning backend for single and multisession i-vector speaker recognition 

      Ghahabi Esfahani, Omid; Hernando Pericás, Francisco Javier (2017-04-01)
      Article
      Open Access
      The lack of labeled background data makes a big performance gap between cosine and Probabilistic Linear Discriminant Analysis (PLDA) scoring baseline techniques for i-vectors in speaker recognition. Although there are some ...
    • Deep neural networks for i-vector language identification of short utterances in cars 

      Ghahabi Esfahani, Omid; Bonafonte Cávez, Antonio; Hernando Pericás, Francisco Javier; Moreno Bilbao, M. Asunción (International Speech Communication Association (ISCA), 2016)
      Conference report
      Restricted access - publisher's policy
      This paper is focused on the application of the Language Identification (LID) technology for intelligent vehicles. We cope with short sentences or words spoken in moving cars in four languages: English, Spanish, German, ...
    • Detection and handling of overlapping speech for speaker diarization 

      Zelenak, Martin; Hernando Pericás, Francisco Javier (2012)
      Conference report
      Open Access
      This thesis concerns the detection of overlapping speech segments and its further application for the improvement of speaker diarization performance. We propose the use of three spatial cross-correlation-based parameters ...
    • Discriminación robusta de locutores 

      Hernando Pericás, Francisco Javier (1996)
      Conference report
      Open Access
    • Discriminative weighting of dynamic feautres in continuous-density hidden Markov models for word recognition 

      Hernando Pericás, Francisco Javier (1995)
      Conference report
      Open Access
      Speech dynamic features, which provide smoothed estimates of the derivatives of the spectral parameter trajectories in the current frame, are routinely used in current speech recognition systems in combination with short-term ...
    • DNN speaker embeddings using autoencoder pre-training 

      Khan, Umair; Hernando Pericás, Francisco Javier (Institute of Electrical and Electronics Engineers (IEEE), 2019)
      Conference lecture
      Restricted access - publisher's policy
      Over the last years, i-vectors have been the state-of-the-art approach in speaker recognition. Recent improvements in deep learning have increased the discriminative quality of i-vectors. However, deep learning architectures ...
    • Double multi-head attention for speaker verification 

      India Massana, Miquel Àngel; Safari, Pooyan; Hernando Pericás, Francisco Javier (Institute of Electrical and Electronics Engineers (IEEE), 2021)
      Conference report
      Open Access
      Most state-of-the-art Deep Learning systems for text-independent speaker verification are based on speaker embedding extractors. These architectures are commonly composed of a feature extractor front-end together with a ...
    • Dynamic time warping applied to detection of confusable word pairs in automatic speech recognition 

      Anguita Ortega, Jan; Hernando Pericás, Francisco Javier (Escola Tècnica Superior d'Enginyers de Telecomunicació de Barcelona, 2005)
      Article
      Open Access
      In this paper we present a rnethod to predict if two words are likely to be confused by an Autornatic SpeechRecognition (ASR) systern. This method is based on the c1assical Dynamic Time Warping (DTW) technique. This ...
    • End-to-end transparent user identification using touchscreen biometrics 

      Krzeminski, Michal; Hernando Pericás, Francisco Javier (Universidad de Málaga, 2020)
      Conference report
      Open Access
      We study the touchscreen data as behavioral biometrics. The goal was to create an end-to-end system that can transparently identify users using raw data from mobile devices. The touchscreen biometrics was researched only ...
    • Esquema unificado de parametrización de la señal de voz en reconocimiento del habla 

      Hernando Pericás, Francisco Javier; Nadeu Camprubí, Climent; Vallverdú Bayés, Sisco (Universidad de Valladolid, 1995)
      Conference report
      Open Access
      A correct choice of voice signal modeling methods is essential to obtain good results in automatic speech recognition. In this paper, we have proposed a unified view of the speech parametrization stage, in which conventional ...
    • Estudio comparativo y nuevas propuestas de tecnicas de parametrizacion de la señal de voz para el reconocimiento del habla 

      Clot, J; Hernando Pericás, Francisco Javier; Nadeu Camprubí, Climent (1994)
      Conference report
      Open Access
      A correct choice of voice signal modeling method is essential to obtain good results in automatic speech recogniton. In this paper, a comparative study betwen two speech signal models, Linear Prediction Coeficients and ...
    • Examen Final 

      Oliveras Vergés, Albert; Hernando Pericás, Francisco Javier (Universitat Politècnica de Catalunya, 2013-01-17)
      Exam
      Restricted access to the UPC academic community
    • Examen Final 

      Oliveras Vergés, Albert; Mariño Acebal, José Bernardo; Hernando Pericás, Francisco Javier; Villares Piera, Nemesio Javier (Universitat Politècnica de Catalunya, 2014-06-10)
      Exam
      Restricted access to the UPC academic community
    • Feature classification by means of Deep Belief Networks for speaker recognition 

      Safari, Pooyan; Ghahabi Esfahani, Omid; Hernando Pericás, Francisco Javier (Institute of Electrical and Electronics Engineers (IEEE), 2015)
      Conference report
      Restricted access - publisher's policy
      In this paper, we propose to discriminatively model target and impostor spectral features using Deep Belief Networks (DBNs) for speaker recognition. In the feature level, the number of impostor samples is considerably ...
    • Frequency and time filtering of filter-bank energies for HMM speech recognition 

      Nadeu Camprubí, Climent; Mariño Acebal, José Bernardo; Hernando Pericás, Francisco Javier; Nogueiras Rodríguez, Albino (H. TIMOTHY BRUMMELL, WILLIAM IDSARDI CITATION DELAWARE, NEW CASTLE, DELAWARE, 1996)
      Conference lecture
      Restricted access - publisher's policy
      In speech recognition, a discriminative frequency weighting can be achieved by decorrelating the frequency sequence of log mel-scaled filter-bank energies with a computationally inexpensive filter. We show how the spectral ...
    • From features to speaker vectors by means of restricted Boltzmann machine adaptation 

      Safari, Pooyan; Ghahabi Esfahani, Omid; Hernando Pericás, Francisco Javier (2016)
      Conference lecture
      Open Access
      Restricted Boltzmann Machines (RBMs) have shown success in different stages of speaker recognition systems. In this paper, we propose a novel framework to produce a vector-based representation for each speaker, which will ...