Eduardo Rodríguez Banga
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Eduardo Rodríguez Banga has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Carmen Magariños
Galician Language Institute, University of Santiago de Compostela, Santiago de Compostela, Spain
6 shared publications Daniel Erro
Ikerbasque, Aholab, University of the Basque Country Ingeniaritza Goi Eskola Teknikoa, Urkijo Zum. z/g, Bilbao, 48013, Spain
6 shared publications- Laura Docio-Fernandez
GTM Research Group, AtlanTTic Research Center, University of Vigo, Maxwell Street, 36310 Vigo, Spain
4 shared publications - Paula Lopez-Otero
Universidade da Coruña - CITIC
4 shared publications F. Campillo
Universidad de Vigo-Signal Theory Group, ETSI Ingeniera de Telecomunicacion, Campus Universitario, Vigo, Pontevedra 36310, Spain
3 shared publicationsFrancisco Campillo Díaz
3 shared publications
Latest Publications
- Open Access
- 18 October 2024
- Published by MDPI AG in Technologies
- Vol. 12 (10) , 205
SWL-LSE: A Dataset of Health-Related Signs in Spanish Sign Language with an ISLR Baseline Method
Progress in automatic sign language recognition and translation has been hindered by the scarcity of datasets available for the training of machine learning algorithms, a challenge that is even more acute for languages with smaller signing communities, such as Spanish. In this paper, we introduce a dataset of 300 isolated signs in Spanish Sign Language, collected online via a web application with contributions from 124 participants, resulting in a total of 8000 instances. This dataset, which is openly available, includes keypoints extracted using MediaPipe Holistic. The goal of this paper is to describe the construction and characteristics of the dataset and to provide a baseline classification method using a spatial–temporal graph convolutional network (ST-GCN) model, encouraging the scientific community to improve upon it. The experimental section offers a comparative analysis of the method’s performance on the new dataset, as well as on two other well-known datasets. The dataset, code, and web app used for data collection are freely available, and the web app can also be used to test classifier performance on-line in real-time.
- 18 October 2024
- Published by MDPI AG in Technologies
- Vol. 12 (10) , 205
Open AccessSWL-LSE: A Dataset of Health-Related Signs in Spanish Sign Language with an ISLR Baseline Method
Progress in automatic sign language recognition and translation has been hindered by the scarcity of datasets available for the training of machine learning algorithms, a challenge that is even more acute for languages with smaller signing communities, such as Spanish. In this paper, we introduce a dataset of 300 isolated signs in Spanish Sign Language, collected online via a web application with contributions from 124 participants, resulting in a total of 8000 instances. This dataset, which is openly available, includes keypoints extracted using MediaPipe Holistic. The goal of this paper is to describe the construction and characteristics of the dataset and to provide a baseline classification method using a spatial–temporal graph convolutional network (ST-GCN) model, encouraging the scientific community to improve upon it. The experimental section offers a comparative analysis of the method’s performance on the new dataset, as well as on two other well-known datasets. The dataset, code, and web app used for data collection are freely available, and the web app can also be used to test classifier performance on-line in real-time.
- 12 December 2018
- Published by Elsevier in Computer Speech & Language
- Vol. 55, 168-186
Language-independent acoustic cloning of HTS voices
- 12 December 2018
- Published by Elsevier in Computer Speech & Language
- Vol. 55, 168-186
Language-independent acoustic cloning of HTS voices
- Open Access
- 1 December 2017
- Published by Institution of Engineering and Technology (IET) in IET Signal Processing
- Vol. 11 (9) , 1023-1030
Influence of speaker de‐identification in depression detection
Depression is a common mental disorder that is usually addressed by outpatient treatments that favour patients’ inclusion in the society. This raises the need for tools to remotely monitor the emotional state of the patients, which can be carried out via telephone or the Internet using speech processing approaches. However, these strategies lead to privacy concerns caused by the transmission of the patients’ speech and its subsequent storage in servers. The use of speech de‐identification to protect the privacy of these patients seems straightforward, but the influence of this procedure in the manifestation of the disease in the patients’ speech has not been addressed yet. Hence, this study evaluates the performance of an automatic depression level estimation system when dealing with original and de‐identified speech, in order to analyse the influence of the de‐identification procedure in the detection of depression. Two de‐identification approaches based on voice transformation via frequency warping and amplitude scaling are assessed, which can be applied to any speaker without additional training. Experiments carried out in the framework of the audio/visual emotion challenge 2014 show that the proposed de‐identification approaches achieve promising de‐identification results at the expense of a slight degradation of depression detection.
- 1 December 2017
- Published by Institution of Engineering and Technology (IET) in IET Signal Processing
- Vol. 11 (9) , 1023-1030
Open AccessInfluence of speaker de‐identification in depression detection
Depression is a common mental disorder that is usually addressed by outpatient treatments that favour patients’ inclusion in the society. This raises the need for tools to remotely monitor the emotional state of the patients, which can be carried out via telephone or the Internet using speech processing approaches. However, these strategies lead to privacy concerns caused by the transmission of the patients’ speech and its subsequent storage in servers. The use of speech de‐identification to protect the privacy of these patients seems straightforward, but the influence of this procedure in the manifestation of the disease in the patients’ speech has not been addressed yet. Hence, this study evaluates the performance of an automatic depression level estimation system when dealing with original and de‐identified speech, in order to analyse the influence of the de‐identification procedure in the detection of depression. Two de‐identification approaches based on voice transformation via frequency warping and amplitude scaling are assessed, which can be applied to any speaker without additional training. Experiments carried out in the framework of the audio/visual emotion challenge 2014 show that the proposed de‐identification approaches achieve promising de‐identification results at the expense of a slight degradation of depression detection.
- 1 November 2017
- Published by Elsevier in Computer Speech & Language
- Vol. 46, 36-52
Reversible speaker de-identification using pre-trained transformation functions
- 1 November 2017
- Published by Elsevier in Computer Speech & Language
- Vol. 46, 36-52
Reversible speaker de-identification using pre-trained transformation functions
- 4 November 2016
- Published by Springer Nature
- p. 54-63
Language-Independent Acoustic Cloning of HTS Voices: An Objective Evaluation
- 4 November 2016
- Published by Springer Nature
- p. 54-63
Language-Independent Acoustic Cloning of HTS Voices: An Objective Evaluation
- 1 July 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 32, 1-5
Piecewise linear definition of transformation functions for speaker de-identification
The main drawback of speaker de-identification approaches using voice conversion techniques is the need for parallel corpora to train transformation functions between the source and target speakers. In this paper, a voice conversion approach that does not require training any parameters is proposed: it consists in manually defining frequency warping (FW) based transformations by using piecewise linear approximations. An analysis of the de-identification capabilities of the proposed approach using FW only or combined with FW modification and spectral amplitude scaling (AS) was performed. Experimental results show that, using the manually defined transformations using only FW, it is not possible to obtain de-identified natural sounding speech. Nevertheless, when modifying the FW, both de-identification accuracy and naturalness increase to a great extent. A slight improvement in de-identification was also obtained when applying spectral amplitude scaling.
- 1 July 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 32, 1-5
Piecewise linear definition of transformation functions for speaker de-identification
The main drawback of speaker de-identification approaches using voice conversion techniques is the need for parallel corpora to train transformation functions between the source and target speakers. In this paper, a voice conversion approach that does not require training any parameters is proposed: it consists in manually defining frequency warping (FW) based transformations by using piecewise linear approximations. An analysis of the de-identification capabilities of the proposed approach using FW only or combined with FW modification and spectral amplitude scaling (AS) was performed. Experimental results show that, using the manually defined transformations using only FW, it is not possible to obtain de-identified natural sounding speech. Nevertheless, when modifying the FW, both de-identification accuracy and naturalness increase to a great extent. A slight improvement in de-identification was also obtained when applying spectral amplitude scaling.
- 1 March 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. e85 d, 5615-5619
Language-independent acoustic cloning of HTS voices: A preliminary study
This paper presents a new method for cross-lingual speaker adaptation in the framework of HMM-based speech synthesis. Taking two HTS voice models as input, one for the desired language and another for the aimed speaker identity, it yields a third model that produces speech in the target language while sounding like the target speaker. The method operates at segmental level (spectral information and average fundamental frequency) and does not require any phonetic or linguistic information. Perceptual evaluation experiments show that, when the input models are good enough, the resulting synthetic voice is perceived as similar to the target speaker with no important quality degradation.
- 1 March 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. e85 d, 5615-5619
Language-independent acoustic cloning of HTS voices: A preliminary study
This paper presents a new method for cross-lingual speaker adaptation in the framework of HMM-based speech synthesis. Taking two HTS voice models as input, one for the desired language and another for the aimed speaker identity, it yields a third model that produces speech in the target language while sounding like the target speaker. The method operates at segmental level (spectral information and average fundamental frequency) and does not require any phonetic or linguistic information. Perceptual evaluation experiments show that, when the input models are good enough, the resulting synthetic voice is perceived as similar to the target speaker with no important quality degradation.
- 4 August 2011
- Published by Institution of Engineering and Technology (IET) in Electronics Letters
- Vol. 47 (16) , 937-938
Multiple f 0 contour parallel Viterbi search for unit selection speech synthesis
Unit selection speech systems generate synthetic speech by concatenation of acoustic units that are chosen by means of a Viterbi search on a previously recorded corpus of the same speaker. Given the finite nature of the corpus, there is usually a trade-off between the acoustic quality of the signal, and the quality of the prosody. Proposed is a parallel Viterbi search considering multiple intonation contours that improves at the same time both acoustic quality and intonation, while maintaining the computational efficiency of the process.
- 4 August 2011
- Published by Institution of Engineering and Technology (IET) in Electronics Letters
- Vol. 47 (16) , 937-938
Multiple f 0 contour parallel Viterbi search for unit selection speech synthesis
Unit selection speech systems generate synthetic speech by concatenation of acoustic units that are chosen by means of a Viterbi search on a previously recorded corpus of the same speaker. Given the finite nature of the corpus, there is usually a trade-off between the acoustic quality of the signal, and the quality of the prosody. Proposed is a parallel Viterbi search considering multiple intonation contours that improves at the same time both acoustic quality and intonation, while maintaining the computational efficiency of the process.
- 28 April 2011
- Published by Institution of Engineering and Technology (IET) in Electronics Letters
- Vol. 47 (9) , 569-570
Segmentwise unit selection
Unit selection speech systems generate synthetic speech by concatenation of acoustic units extracted from a natural recording. Given a large speech database, the sequence of units with the best global cost is chosen by means of a Viterbi search. In this reported work, it is shown that small subcosts not related to perceptual measures can affect the sequence of units that is finally chosen, with a potential effect on the quality of synthetic speech. A segmentwise unit selection approach that minimises this effect is then proposed.
- 28 April 2011
- Published by Institution of Engineering and Technology (IET) in Electronics Letters
- Vol. 47 (9) , 569-570
Segmentwise unit selection
Unit selection speech systems generate synthetic speech by concatenation of acoustic units extracted from a natural recording. Given a large speech database, the sequence of units with the best global cost is chosen by means of a Viterbi search. In this reported work, it is shown that small subcosts not related to perceptual measures can affect the sequence of units that is finally chosen, with a potential effect on the quality of synthetic speech. A segmentwise unit selection approach that minimises this effect is then proposed.
- Open Access
- 1 September 2010
- Published by MIT Press in Computational Linguistics
- Vol. 36 (3) , 505-533
Hierarchical Phrase-Based Translation with Weighted Finite-State Transducers and Shallow-n Grammars
In this article we describe HiFST, a lattice-based decoder for hierarchical phrase-based translation and alignment. The decoder is implemented with standard Weighted Finite-State Transducer (WFST) operations as an alternative to the well-known cube pruning procedure. We find that the use of WFSTs rather than k-best lists requires less pruning in translation search, resulting in fewer search errors, better parameter optimization, and improved translation performance. The direct generation of translation lattices in the target language can improve subsequent rescoring procedures, yielding further gains when applying long-span language models and Minimum Bayes Risk decoding. We also provide insights as to how to control the size of the search space defined by hierarchical rules. We show that shallow-n grammars, low-level rule catenation, and other search constraints can help to match the power of the translation system to specific language pairs.
- 1 September 2010
- Published by MIT Press in Computational Linguistics
- Vol. 36 (3) , 505-533
Open AccessHierarchical Phrase-Based Translation with Weighted Finite-State Transducers and Shallow-n Grammars
In this article we describe HiFST, a lattice-based decoder for hierarchical phrase-based translation and alignment. The decoder is implemented with standard Weighted Finite-State Transducer (WFST) operations as an alternative to the well-known cube pruning procedure. We find that the use of WFSTs rather than k-best lists requires less pruning in translation search, resulting in fewer search errors, better parameter optimization, and improved translation performance. The direct generation of translation lattices in the target language can improve subsequent rescoring procedures, yielding further gains when applying long-span language models and Minimum Bayes Risk decoding. We also provide insights as to how to control the size of the search space defined by hierarchical rules. We show that shallow-n grammars, low-level rule catenation, and other search constraints can help to match the power of the translation system to specific language pairs.