RiuNet UPV Home
:
Browsing by Author

Search RiuNet

Browse

All of RiuNet

My Account

RiuNet Help

Admin. UPV

Browsing by Author "Giménez Pastor, Adrián"

Now showing items 1-20 of 28

A comparative study of Neural Machine Translation frameworks for the automatic translation of open data resources

Iranzo Sánchez, Javier (Universitat Politècnica de València, 2018-09-11)

[EN] Machine Translation (MT) is one of the most active areas in Artificial Intelligence, particularly in Pattern Recognition. MT has recently received a great deal of attention by key technology players such as Google, ...
Adaptación y evaluación de modelos de diarización de locutores para recursos educativos abiertos

Ciscar Martinez, Vicent Andreu (Universitat Politècnica de València, 2019-01-14)

La Educación Abierta se ha convertido en una aproximación revolucionaria para el futuro de la educación permitiendo el acceso mundial a un gran volumen de Recursos Educativos Abiertos (REA). Un ejemplo emblemático de REA ...
Albayzin Evaluation: The PRHLT-UPV Audio Segmentation System

Silvestre Cerdà, Joan Albert; Giménez Pastor, Adrián; Andrés Ferrer, Jesús; Civera Saiz, Jorge; Juan Císcar, Alfonso (Ramos Castro, Daniel, 2012-11-21)

This paper describes the audio segmentation system developed by the PRHLT research group at the UPV for the Albayzin Audio Segmentation Evaluation 2012. The PRHLT-UPV audio segmentation system is based on a conventional ...
Arabic Printed Word Recognition Using Windowed Bernoulli HMMs

Alkhoury, Ihab; Giménez Pastor, Adrián; Juan Císcar, Alfonso; Andrés Ferrer, Jesús (Springer Verlag, 2013-09-09)

[EN] Hidden Markov Models (HMMs) are now widely used for off-line text recognition in many languages and, in particular, Arabic. In previous work, we proposed to directly use columns of raw, binary image pixels, which are ...
Bernoulli HMMs at subword level for handwritten word recognition

Giménez Pastor, Adrián; Juan, Alfons (Springer Verlag (Germany), 2009)

This paper presents a handwritten word recogniser based on HMMs at subword level (characters) in which state-emission probabilities are governed by multivariate Bernoulli probability functions. This recogniser works directly ...
Bernoulli HMMs for Handwritten Text Recognition

Giménez Pastor, Adrián (Universitat Politècnica de València, 2014-06-09)

In last years Hidden Markov Models (HMMs) have received significant attention in the task off-line handwritten text recognition (HTR). As in automatic speech recognition (ASR), HMMs are used to model the probability of ...
Comparació de les eines informàtiques TLK i Kaldi per al desenvolupament de sistemes de reconeixement de la parla en català/valencià

Baquero Arnal, Pau (Universitat Politècnica de València, 2016-09-05)

[CA] En aquest treball es comparen les eines de reconeixement de la parla TLK i Kaldi per al desenvolupament de sistemes de reeconeixement en catalá.
Comparison of Bernoulli and Gaussian HMMs using a vertical repositioning technique for off-line handwriting recognition

Doetsch, Patrick; Hamdani, Mahdi; Ney, Hermann; Giménez Pastor, Adrián; Andrés Ferrer, Jesús; Juan Císcar, Alfonso (Institute of Electrical and Electronics Engineers (IEEE), 2012-09-18)

—In this paper a vertical repositioning method based on the center of gravity is investigated for handwriting recognition systems and evaluated on databases containing Arabic and French handwriting. Experiments show ...
Comparison of Data Selection Techniques for the Translation of Video Lectures

Wuebker, Joern; Ney, Hermann; Martínez-Villaronga, Adrià; Giménez Pastor, Adrián; Juan Císcar, Alfonso; Servan, Christophe; Dymetman, Marc; Mirkin, Shashar (Association for Machine Translation in the Americas, 2014-10-22)

[EN] For the task of online translation of scientific video lectures, using huge models is not possible. In order to get smaller and efficient models, we perform data selection. In this paper, we perform a qualitative ...
Development and Evaluation of an Automatic Speech Recognition System Adapted to the Transcription of Classroom Video Recordings

Roselló Beneitez, Nahuel Unai (Universitat Politècnica de València, 2021-10-14)

[ES] El Reconocimiento Automático del Habla (RAH) ha demostrado ser una manera efectiva y eficiente de convertir habla a texto a lo largo de los últimos años. Este trabajo, desarrollado en el contexto de dos proyectos ...
Direct Segmentation Models for Streaming Speech Translation

Iranzo-Sánchez, Javier; Giménez Pastor, Adrián; Silvestre Cerdà, Joan Albert; Baquero-Arnal, Pau; Civera Saiz, Jorge; Juan, Alfons (Association for Computational Linguistics, 2020-11-20)

[EN] The cascade approach to Speech Translation (ST) is based on a pipeline that concatenates an Automatic Speech Recognition (ASR) system followed by a Machine Translation (MT) system. These systems are usually connected by ...
Discriminative Bernoulli HMMs for isolated handwritten word recognition

Giménez Pastor, Adrián; Andrés Ferrer, Jesús; Juan, Alfons (Elsevier, 2014-01-01)

[EN] Bernoulli HMMs (BHMMs) have been successfully applied to handwritten text recognition (HTR) tasks such as continuous and isolated handwritten words. BHMMs belong to the generative model family and, hence, are usually ...
Discriminative Bernoulli Mixture Models for Handwritten Digit Recognition

Giménez Pastor, Adrián; Andrés Ferrer, Jesús; Juan Císcar, Alfonso; Serrano Martinez Santos, Nicolas (Institute of Electrical and Electronics Engineers (IEEE), 2011-09-18)

Bernoulli-based models such as Bernoulli mixtures or Bernoulli HMMs (BHMMs), have been successfully applied to several handwritten text recognition (HTR) tasks which range from character recognition to continuous and ...
Handwriting word recognition using windowed Bernoulli HMMs

Giménez Pastor, Adrián; Alkhoury, Ihab; Andrés Ferrer, Jesús; Juan Císcar, Alfonso (Elsevier, 2014-01)

[EN] Hidden Markov Models (HMMs) are now widely used for off-line handwriting recognition in many lan- guages. As in speech recognition, they are usually built from shared, embedded HMMs at symbol level, where state-conditional ...
Interactive handwriting recognition with limited user effort

Serrano Martinez Santos, Nicolas; Giménez Pastor, Adrián; Civera Saiz, Jorge; Sanchis Navarro, José Alberto; Juan Císcar, Alfonso (Springer Verlag (Germany), 2014-03-01)

[EN] Transcription of handwritten text in (old) documents is an important, time-consuming task for digital libraries. Although post-editing automatic recognition of handwritten text is feasible, it is not clearly better ...
Language technology for handwritten text recognition

Toselli, Alejandro Héctor; Serrano Martínez-Santos, Nicolás; Giménez Pastor, Adrián; Khoury, Ihab; Juan Císcar, Alfonso; Vidal Ruiz, Enrique (Springer Verlag (Germany), 2012)

This paper shows how the nowadays prevalent technology used in HTR borrows concepts and methods from the field of ASR; i.e. those based on Hidden Markov Models (HMMs). Additionally, it will be described a HTR approach based ...
Live Streaming Speech Recognition Using Deep Bidirectional LSTM Acoustic Models and Interpolated Language Models

Jorge-Cano, Javier; Giménez Pastor, Adrián; Silvestre Cerdà, Joan Albert; Civera Saiz, Jorge; Sanchis Navarro, José Alberto; Juan, Alfons (Institute of Electrical and Electronics Engineers, 2022)

[EN] Although Long-Short Term Memory (LSTM) networks and deep Transformers are now extensively used in offline ASR, it is unclear how best offline systems can be adapted to work with them under the streaming setup. After ...
Online Multilingual Neural Machine Translation

Iranzo Sánchez, Javier (Universitat Politècnica de València, 2019-10-30)

[ES] La traducción automática (MT) es un área de investigación sobre el desarrollo de sistemas que traducen textos de manera automática. Actualmente, la mayoría de sistemas de MT utilizan redes neuronales en lo que se ...
Reconeixement de paraules manuscrites amb HMMs amb mixtures de Bernoulli i de gaussianes als

Giménez Pastor, Adrián (Universitat Politècnica de València, 2011-10-18)

En este trabajo presentamos un sistema de reconocimiento de palabras manuscritas basado en HMMs segmentados con mixturas de Bernoulli en los estados. Este sistema es aplicado al corpus IAM y comparado con otro, basado ...
Speaker-Adapted Confidence Measures for ASR using Deep Bidirectional Recurrent Neural Networks

Del Agua Teba, Miguel Angel; Giménez Pastor, Adrián; Sanchis Navarro, José Alberto; Civera Saiz, Jorge; Juan, Alfons (Institute of Electrical and Electronics Engineers, 2018)

[EN] In the last years, Deep Bidirectional Recurrent Neural Networks (DBRNN) and DBRNN with Long Short-Term Memory cells (DBLSTM) have outperformed the most accurate classifiers for confidence estimation in automatic speech ...