Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- José Manuel PardoSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works132 from public data
- Error Analysis of Statistical Machine Translation Output337
A framework for classification of the errors of a machine translation system is presented and an error analysis of the system used by the RWTH in the first TC-STAR evaluation is carried out.
- Classification of epileptic EEG recordings using signal transforms and convolutional neural networks172
This analysis was carried out using two public datasets (Bern-Barcelona EEG and Epileptic Seizure Recognition datasets) obtaining significant improvements in accuracy.
- The Fourth Dialog State Tracking Challenge134
This edition of the fourth dialog state tracking challenge again focused on human-human dialogs, but introduced the task of cross-lingual adaptation of trackers, which received a total of 32 entries from 9 research groups.
- Speech to sign language translation system for Spanish119
The development of and the first experiments in a Spanish to sign language translation system in a real domain focusing on the sentences spoken by an official when assisting people applying for, or renewing their Identity Card are described.
- 84
- DynaEval: Unifying Turn and Dialogue Level Evaluation82
DynaEval, a unified automatic evaluation framework which is not only capable of performing turn-level evaluation, but also holistically considers the quality of the entire dialogue, is proposed.
- Overview of the Ninth Dialog System Technology Challenge: DSTC973
The task definition is described, provided datasets, baselines, and evaluation setup for each track, and the results of the submitted systems are summarized to highlight the general trends of the state-of-the-art technologies for the tasks.
- 71
- RevUP: Automatic Gap-Fill Question Generation from Educational Texts64
RevUP fills the semantic gap left open by previous work in this area, and represents a significant step towards automatically generating quality tests for teachers and self-motivated learners.
- 54
- 47
- 43
- A Web-based Platform for Collection of Human-Chatbot Interactions39
An advanced platform for evaluating and annotating human-chatbot interactions, its main features and goals, as well as the future plans the authors have for it are described.
- Frequency features and GMM-UBM approach for gait-based person identification using smartphone inertial signals38
The final results demonstrate the capability of the GMM-UBM approach for gait recognition, and show how the PLP coefficients can improve system performance while reducing drastically the number of features.
- Phonotactic language recognition using ivvectors and phoneme posteriogram counts38
This paper describes a novel approach to phonotactic LID, where instead of using soft-counts based on phoneme lattices, the high-dimensional vectors of counts are reduced to low-dimensional units for which the commonly used term i-vectors is adapted.
- Phonotactic language recognition using i-vectors and phoneme posteriogram counts38
This paper describes a novel approach to phonotactic LID, where instead of using soft-counts based on phoneme lattices, the high-dimensional vectors of counts are reduced to low-dimensional units for which the commonly used term i-vectors is adapted.
- Automatic Correction of ASR Outputs by Using Machine Translation37
This paper proposes an automatic correction procedure based on using a phrase-based machine translation system trained using words and phonetic encoding representations to the generated n-best lists of ASR results, which can provide a quick and robust mechanism to improve the performance of the ASR.
- A Spanish speech to sign language translation system for assisting deaf-mute people37
The developed system translates officer explanations into sign language for deafmute people when assisting people in applying for, or renewing the National Identification Document and the Passport is focused on the sentences spoken by an officer.
- 36
- Extended phone log-likelihood ratio features and acoustic-based i-vectors for language recognition32
This paper presents new techniques with relevant improvements added to the primary system presented by the group to the Albayzin 2012 LRE competition, where the use of any additional corpora for training or optimizing the models was forbidden.
- 31
- 30
- 28
- Patrol team language identification system for DARPA RATS P1 evaluation28
This paper describes the language identification (LID) system developed by the Patrol team for the first phase of the DARPA RATS (Robust Automatic Transcription of Speech) program, which seeks to advance state of the art detection capabilities on audio from highly degraded communication channels.
- D-Score: Holistic Dialogue Evaluation Without Reference27
An automatic dialogue evaluation framework called D-score is proposed that resembles the way gymnastics is evaluated and learns to appreciate the overall quality of human-human conversations through a representation that is shared by all tasks without over-fitting to individual task domain.
- 27
- Overview of the seventh Dialog System Technology Challenge: DSTC725
This paper summarizes the overall setup and results of DSTC7, including detailed descriptions of the different tracks, provided datasets and annotations, overview of the submitted systems and their final results, and discussed the future directions for the challenge (DSTC8).
- MDD-Eval: Self-Training on Augmented Data for Multi-Domain Dialogue Evaluation24
The proposed MDD-Eval framework first train a teacher evaluator with human-annotated data to acquire a rating skill to tell good dialogue responses from bad ones in a particular domain and then, adopt a self-training strategy to train a new evaluators with teacher-annotation multi-domain data, that helps the newevaluator to generalize across multiple domains.
- Report from the NSF Future Directions Workshop on Automatic Evaluation of Dialog: Research Directions and Challenges23
The workshop explored the current state of the art along with its limitations and suggested promising directions for future work in this important and very rapidly changing area of research.
- Evaluation of the Neurological State of People with Parkinson’s Disease Using i-Vectors22
The suitability of the i-vector approach to monitor the neurological state of people with Parkinson’s Disease is shown.
- SARA: Singapore’s Automated Responsive Assistant, A Multimodal Dialogue System for Touristic Information21
SARA, a multimodal dialogue system offering touristic assistance for visitors coming to Singapore, is implemented as an Android mobile phone application and provides information about local attractions, restaurants, sightseeing, direction and transportation services.
- A speech interface for air traffic control terminals21
Both objective and subjective performance obtained in field evaluation of the speech interface are presented, encouraging future development and integration phases in the Spanish ATC terminals designed by Aeropuertos Espanoles y Navegacion Aerea.
- 20
- Design and evaluation of a conversational agent for the touristic domain20
The design and evaluation of SARA, a conversational agent for the touristic domain featuring a high number of different, unique characteristics: spoken dialogue interaction, dialogue orchestration, context dependent information, an animated avatar and support for different kind of dialogue types.
- A Multimodal Interface for Access to Content in the Home20
The application of multimodal interface technologies to searching and browsing a database of movies and the resulting system allows users to access movies using speech, pen, remote control, and dynamic combinations of these modalities.
- 20
- Multimodal Audio-Language Model for Speech Emotion Recognition19
This paper combines an audio encoder, specifically the Whisper-large-v3 model, with LLMs Phi 1.5 and Gemma 2b to create a robust and effective system for categorizing emotions in speech.
- Automatic evaluation of end-to-end dialog systems with adequacy-fluency metrics19
A two-dimensional evaluation metric that is designed to operate at sentence level, which considers the syntactic and semantic information carried along the answers generated by an end-to-end dialog system with respect to a set of references is evaluated.
- CLARA: A Multifunctional Virtual Agent for Conference Support and Touristic Information19
A multifunctional conversational agent which combines natural language search capabilities for two different domain applications: a conference information system and local tourist guide is presented.
- 18
- Revisiting scenarios and methods for variable frame rate analysis in automatic speech recognition18
From the evaluation on a difficult telephone large vocabulary task, it is established that VFR analysis does not significantly improve the results obtained using the traditional fixed frame rate analysis (FFR), except when additive noise is present in the database and specially for low SNRs.
- 17
- Speech into Sign Language Statistical Translation System for Deaf People17
A set of experiments used to develop a statistical system from translating speech to sign language for deaf people is presented, composed of an Automatic Speech Recognition system, followed by a statistical translation module and an animated agent that represents the different signs.
- Overview of Robust and Multilingual Automatic Evaluation Metrics for Open-Domain Dialogue Systems at DSTC 11 Track 416
The datasets and baselines provided to participants are described and the submission and result details of the two proposed subtasks are discussed, which promote robust and multilingual automatic evaluation metrics.
- Language identification based on n-gram frequency ranking16
A novel approach for language identification based on a text categorization technique, namely an n-gram frequency ranking, which overcomes PPRLM due to the inclusion of 4gram and 5-gram in the classifier.
- Evaluation of alternatives on speech to sign language translation16
Three alternative techniques have been evaluated: a rule-based approach, a phrase-based statistical approach, and a approach that makes use of stochastic finite state transducers on speech to sign language machine translation.
- New word-level and sentence-level confidence scoring using graph theory calculus and its evaluation on speech understanding16
This work presents new confidence measures employing the graph theory that shows how to estimate some interesting characteristics about the different paths through the graph that constitute the recognition solutions, without the need of expanding them all.
- 15
- Semi-Automatic Generation of Dialogue Applications in the GEMINI Project15
The development of a flexible platform able to produce user-friendly interactive multilingual and multi-modal dialogue interfaces to databases with a minimum of human effort and the demonstration of the platform’s efficiency through the development of two different applications based on this platform.
- 14
- Automated Scoring of Chatbot Responses in Conversational Dialogue14
This paper proposes a novel approach of combining the word representations of human and chatbot responses, and using machine learning algorithms, such as support vector machines (SVM), random forests (RF), and neural networks (NN), to learn the quality of the chat bot responses.
- Low-resource language recognition using a fusion of phoneme posteriorgram counts, acoustic and glottal-based i-vectors14
The primary system was the fusion of three different i-vector based systems: one acoustic system based on MFCCs, a phonotactic system using trigrams of phone-posteriorgram counts, and another acoustic system using RPLPs that improved robustness against noise.
- 13
- On the use of phone-gram units in recurrent neural networks for language identification13
The phonotactic system performs ~13% better than the unigram-based RNNLM system and is calibrated and fused with other scores from an acoustic-based i-vector system and a traditional PPRLM system, providing additional improvements showing that they provide complementary information to the LID system.
- Language model adaptation for a speech to sign language translation system using web frequencies and a MAP framework13
The technique uses information from the source language, Spanish in the authors' task, and from the phrase-based translation matrix in order to create a new LM, estimated using web frequencies, which adapts the counts of the SLM through the Maximum A Posteriori method (MAP).
- Language identification techniques based on full recognition in an air traffic control task13
The work done in language identification for an air traffic control speech recognizer for continuous speech is presented and it is confirmed that PPRLM is quite sensible to those problems and that a technique based on a Bayesian classifier is the one with the best performance in spite of its higher computational cost.
- PoE: A Panel of Experts for Generalized Automatic Dialogue Assessment12
A Panel of Experts (PoE) network is proposed, a multitask network that consists of a shared transformer encoder and a collection of lightweight adapters that exhibits better zero-shot generalization than existing state-of-the-art ADEMs and the ability to easily adapt to new domains with few-shot transfer learning.
- An advanced platform to speed up the design of multilingual dialog applications for multiple modalities12
This paper presents a complete platform for the semiautomatic and simultaneous generation of human–machine dialog applications in two different and separate modalities and several languages to provide services oriented to obtaining or modifying the information from a database.
- The GEMINI platform: semi-automatic generation of dialogue applications12
The development and implementation of a platform able to produce user-friendly interactive multilingual and multi-modal dialogue interfaces to databases with a minimum of human effort and the demonstration of the platform’s efficiency through the development of two different applications using this platform.
- Automatic Understanding of ATC Speech: Study of Prospectives and Field Experiments for Several Controller Positions11
A system able to identify the language spoken and recognize and understand sentences in both Spanish and English is developed and the first time that field ATC speech is captured, processed, and analyzed.
- 11
- 10
- Automatic Detection of Inconsistencies and Hierarchical Topic Classification for Open-Domain Chatbots10
A novel methodology consisting of two main phases: hierarchical automatic detection of topics and subtopics in dialogue interactions using a zero-shot learning approach, and detecting inconsistent answers using k-means and the Silhouette coefficient is proposed.
- 10
- 10
- Overview of the Tenth Dialog System Technology Challenge: DSTC109
The task definition, provided datasets, baselines, and evaluation setup for each track, and the results of the submitted systems are summarized to highlight the general trends of the state-of-the-art technologies for the tasks.
- Exploring Transformer-based Language Recognition using Phonotactic Information8
The proposed encoder-only approach provides significant improvements, outperforming phonotactic-based RNNs and Glove-based i-Vectors architectures, getting a relative improvement of 5.5% and 38.
- Speeding up the design of dialogue applications by using database contents and structure information8
The main proposed strategies are: the generation of automatic state proposals for defining the dialogue flow network, the automatic selection of slots to be requested using mixed-initiative, the semi-automatic generation of SQL statements, and the quick generation of the data model of the application and the connection with the database fields.
- 8
- Evaluating emotional and subjective responses in synthetic art-related dialogues: A multi-stage framework with large language models7
The findings reveal that the capacity to recover templates or profiles is more effective for information or profile items that are objective and factual, in contrast to those related to mental states or subjective facts.
- Subjective Annotation and Evaluation of Three Different Chatbots WOCHAT: Shared Task Report7
This paper evaluates the performance of three different chatbots: IRIS, TickTock and Joker, that have been made available to the public online to confirm the robustness, scalability and usability of the systems, as well as to detect a few areas in which response accuracy was lacking.
- 7
- 7
- 7
- Strategies to reduce design time in multimodal/multilingual dialog applications7
This paper presents a complete platform for the semiautomatic generation of human-machine dialog systems, that generates dialogs to access the service data in different languages and two modalities, speech and web, simultaneously.
- Investigating the Impact of Pre-trained Language Models on Dialog Evaluation6
This study serves as the first comprehensive assessment of the effects of different Pr-LMs on automatic dialog evaluation, examining their impact on three typical automatic dialog Evaluation metrics across three different dialog evaluation benchmarks.
- Phonotactic Language Recognition Using A Universal Phoneme Recognizer and A Transformer Architecture6
- 6
- 6
- A Discriminative Text Categorization Technique for Language Identification built into a PPRLM System6
A state-of-the-art language identification system based on a parallel phone recognizer, the same as in PPRLM, but instead of using as phonotactic constraints traditional n-gram language models, a new language model is created using a ranking with the most frequent and discriminative n- grams between languages.
- Attention-based Semantic Priming for Slot-filling5
It is proposed that an attention-based RNN architecture can be used to simulate semantic priming for sequence labelling, and pre-trained word embeddings are employed to characterize the semantic relationship between utterances and labels.
- 5
- 5
- Speed Up Strategies for the Creation of Multimodal and Multilingual Dialogue Systems5
This paper summarizes the work done in the PhD thesis and proposes different innovative, dynamic, and intelligent acceleration strategies applied to a development platform for reducing the design time of multimodal and multilingual dialogue systems and to improve the runtime modules.
- Novel Applications of Neural Networks in Speech Technology Systems: Search Space Reduction and Prosodic Modeling5
A double objective is to get accurate predictors for both the fundamental frequency (FO) curve and phoneme duration by minimizing the model estimation error in a Spanish text-to-speech system, by means of a neural network estimator, which has proved to be an excellent tool for the modeling.
- Language identification using several sources of information with a multiple-Gaussian classifier5
To normalize the scores, eliminate the bias in the scores and improve the classifier, the bias removal technique and a Gaussian classifier are compared and it is shown how all improvements have been mostly additive.
- 4
- Unsupervised Mutual Learning of Discourse Parsing and Topic Segmentation in Dialogue4
An unsupervised mutual learning framework (UMLF) that jointly models rhetorical and topic structures, allowing them to mutually reinforce each other without requiring additional annotations is proposed, demonstrating its effectiveness in improving discourse structure modeling.
- 4
- Language recognition using phonotactic-based shifted delta coefficients and multiple phone recognizers4
A new language recognition technique based on the application of the philosophy of the Shifted Delta Coefficients to phone log-likelihood ratio features (PLLR) is described which allows the incorporation of long-span phonetic information at a frame-by-frame level while dealing with the temporal length of each phone unit.
- 3
- Considerations on creating conversational agents for multiple environments and users3
This paper will consider the elements necessary to create a conversational agent for different types of users, environments, and tasks, and give insight on how to scale the proposed solutions.
- 3
- Joint Learning of Word and Label Embeddings for Sequence Labelling in Spoken Language Understanding3
The proposed approach encodes labels using a combination of word embeddings and straightforward word-label association from the training data and can achieve state-of-the-art performance with much fewer trainable parameters.
- 3
- A Multimodal Control Architecture for Autonomous Unmanned Aerial Vehicles3
A multimodal control architecture that enables an operator to manage an autonomous Unmanned Aerial Vehicle through high level tasks in an indoors environment through a web based graphical user interface that can be accessed from any Internet powered device.
- Incorporación de n-gramas discriminativos para mejorar un reconocedor de idioma fonotáctico basado en i-vectores3
Una nueva tecnica que permite combinar la informacion de dos sistemas fonotacticos distintos with el objetivo of mejorar los resultados oficiales de un sistema de reconocimiento automatico de idioma.
- Incorporación de n-gramas discriminativos para mejorar un reconocedor de idioma fonotáctico basado en i-vectores Incorporation of discriminative n-grams to improve a phonotactic language reco gnizer based on i-vectors3
Una nueva tecnica que permite combinar la informacion de dos sistemas fonotacticos distintos with el objetivo of mejorar los resultados oficiales de un sistema de reconocimiento automatico de idioma.
- Implementation of dialog applications in an open-source voiceXML platform3
This paper studies the approach followed to use the VoiceXML standard in a dialog system platform already available in this group, and reviews some relevant modules in this platform and their capabilities, highlighting the use of standards in them.
- 2
- 2
- 2
- Automatic Detection of Inconsistencies in Open-Domain Chatbots2
This paper proposes a new methodology based on three phases: a) automatic detection of dialogue topics using zero-shot learning approaches, b) automatic clustering of distinctive questions, and c) detecting inconsistent answers using K-Means clustering and the Silhouette coefficient.
- Incorporation of Language Discriminative Information into Recurrent Neural Networks Models to LID Tasks2
Considering a phonotactic approach where the phonetic units called “phone-grams” are used, in order to introduce context information in the features used to train the RNNLM, a novel technique is proposed.
- 2
- 2
- Conversational agent and management tools for conference and tourism domain.2
A platform and a set of useful tools that allow the fast creation of conversation agents for two different domain applications: a conference information system and local tourist guide are described.
- The path to success: Failures in rEal Robots (FinE-R)2
The main workshop goal is to provide an open exchange forum to the robotic community where participants can share their personal "failure to success" stories and the papers accepted give a good overview of different types of errors encountered in the robotic fields.
- Perceptive Parallel Processes Coordinating Geometry and Texture.2
The major contribution of this paper consists on the success of combining texture and geometry based solutions running in parallel and sharing information in real time to allow a full generic solution to be able to find almost any present object in a wide environment.
- Design and evaluation of acceleration strategies for speeding up the development of dialog applications2
A complete development platform that features different innovative acceleration strategies that simplify and speed up the definition of the different elements required to design a spoken dialog service and the design time is reduced in more than 56% and the number of keystrokes by 84%.
- New teaching methodology for electronics and its adaptation to the European space for higher education2
The deployment of a new remote access platform tailored to the specific needs of students in a course on digital electronic systems based on microprocessors has enabled them to enjoy greater flexibility by offering them the ability to complete remotely, in a different time and place to the university, the proposed practices (PBL) simultaneously with the lectures.
- Plataforma de generación semiautomática de sistemas de diálogo multimodales y multilingües : proyecto GEMINI2
La plataforma completa de generacion semiautomatica de sistemas de dialogo hombre-maquina se generan simultaneamente dialogos para acceder a los datos de dicho servicio simultaneousamente en varios idiomas y en dos modalidades, voz y web.
- thaulab@EEUCA 2026: Who Said What to Whom? A Targeting-Aware Neural-Symbolic Pipeline for Gaming Toxicity Detection1
A three-stage pipeline combining an ensemble of two compact transformers with a Linguistically-Informed Mediator that resolves inter-model disagreements through corpus-backed lexical normalization, class-conditional unigram scoring, multilingual profanity detection, and agentive targeting analysis grounded in speech act theory is implemented.
- 1
- 1
- 1
- Application of backend database contents and structure to the design of spoken dialog services1
A complete development platform that reduces the design time by using different types of acceleration strategies based on using information from the data model structure and database contents, as well as cumulative information obtained throughout the successive steps in the design.
- A PROPOSAL OF METRICS FOR DETAILED EVALUATION IN PRONUNCIATION MODELING1
This paper is evaluating the effective improvement due to every pronunciation variation used, defining specific improvement metrics on the utterance level and showing how these metrics actually highlight the beneficial impact achieved by the application of phonological rules when dealing with certain pronunciations variants.
- Topic Identification based on Bayesian Belief Networks in the context of an Air Traffic Control Task1
Los resultados obtenidos demuestran la importancia oficial de considerar el orden de aparicion de la informacion y the necesidad de incluirla in las Redes Bayesianas en futuros trabajos.
- –
- –
- DREAM: A Multicultural Multimodal Dataset Linking Dialogues and Realistic Image Sequences–
This work introduces DREAM (Dialogue to REAlistic Multicultural Image Sequences), a multicultural multimodal resource that ties dialogues grounded in explicit persona profiles to photorealistic, storyboard-like image sequences, facilitating research on visually anchored dialogue understanding, consistency, and generation.
- –
- Commonsense Generation and Evaluation for Dialogue Systems using Large Language Models–
Preliminary results suggest that the proposed methodology effectively harnesses LLMs capabilities for commonsense reasoning and evaluation in dialogue systems.
- –
- –
- –
- –
- Strategies for accelerating the design of dialogue applications using heuristic information from the backend database–
This paper describes attempts to take advantage of heuristic information from the contents or structure of the backend database in order to speed up the definition of the dialogue flow using the quick creation of classes and attributes to define the data model structure.
- Aplicación de métodos estadísticos para la traducción de voz a Lengua de Signos–
A set of experiments used to develop a statistical system from translating speech to sign language for deaf people is presented, composed of an Automatic Speech Recognition (ASR) system, followed by a statistical translation module and an animated agent that represents the different signs.
- Adapting a Search Algorithm for the Spanish Railway Network–
A search algorithm adapted to the Spanish Railway Network for generating as many traveling options as possible between two railway stations using Warshall's algorithm, which uses connecting matrices to find all possible railway journeys.
- ANN F0 Modeling for Female-Voice Synthesis in Spanish: restricted and non-restricted domains–
The modeling of the F0 curve of a female voice in several Restricted-Domains and in a general-domain, aimed at developing a Speech Synthesis System for Spanish is described.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.