Belén Ruiz

The BANCA database and evaluation protocol

In this paper we describe the acquisition and content of a new large, realistic and challenging multi-modal database intended for training and testing multi-modal verification systems. The BANCA database was captured in four European languages in two modalities (face and voice). For recording, both high and low quality microphones and cameras were used. The subjects were recorded in three different scenarios, controlled, degraded and adverse over a period of three months. In total 208 people were captured, half men and half women. In this paper we also describe a protocol for evaluating verification algorithms on the database. The database will be made available to the research community through http://www.ee.surrey.ac.uk/Research/VSSP/banca.

Toward an Equal Opportunity Web: Applications, Standards, and Tools that Increase Accessibility

This paper discusses about rich Internet applications and supporting technology, such as JavaScript and Flash. RIAs can offer an interactive experience closer to that of a desktop application.

Design for All in multimedia guides for museums

The Design for All principles define the characteristics which a device should possess in order for it be utilised by every type of user, independent of his or her sensory disabilities or technological competency. These principles are realised in the current work as an integrative tool with which to facilitate universal access to museums via multimedia and portable guides, making access available to all. Based on these principles, this article describes the main findings of the design and use of the MGA (Multimedia Guides for All) approach proposed in this paper. This approach involves a series of recommendations for the selection, application, preparation of content, and maintenance of this type of computerized device, in order to achieve these principles. Firstly, a comparative analysis of the principal types of electronic guides available in museums which incorporate accessibility criteria is provided. Subsequently, a real case study case is presented which conforms to the MGA approach. The MGA approach can be applied to other domains, which should be explored in further research. The conclusions in this article have been drawn from two R&D projects financed by the Spanish Ministry of Industry, Science and Technology supported by the EC FEDER R&D Program.

Inclusion of Accessibility Requirements in the Design of Electronic Guides for Museums

Technology can be an integrative tool with which to facilitate universal access to museums and their works via virtual guides that accompany the user making the visit more inclusive and accessible. With this aim it is necessary to keep in mind accessibility requirements which guide the design and development process. Since there are no clear procedures for defining and including these requirements, this paper presents a proposal for requirements capture and definitions of accessibility mechanisms which assure their integration, following studies and standards in the area of accessibility.

Speech/Speaker Recognition Using a HMM/GMM Hybrid Model

In this paper, a speaker recognition voice based system is presented [5]. We have implemented it in a Sun platform.We train (and test) the system using a Database recorded in several sessions in order to repair the huge effects that the speech variability with time has in the recognition rate system. Several experiments have been made in order to achieve the best configuration in the system set up. This is an important point to take into account in a real world system in which users train the system once and the models generated in the training process are not updated for strategic reasons. The recognition rate obtained for the proposed system is around 93% if the speech came from a microphone is around 90% when the speech came from a phone line.

Real-time subtitle synchronization in live television programs

The number of live programs in broadcast and IPTV Digital Television is increasing as the offer of pre-recorded multimedia in Internet television and video on-demand grows. For many people, subtitles are the only way to access the audio of television programs and in live subtitling environments the audio transcription process will deliver the corresponding subtitles several seconds after the speech fragment has been received. The disturbing effect of the resulting lack of synchronism is still a challenging issue in the accessibility to live-multimedia. This paper presents the application of an Automatic Speech Recognition and synchronizing system, tailored for three different live television environments (i.e. news programs, sports, magazine). Individual subtitle delays are measured and used to synchronize video/audio and subtitles before delivering the TV program in a dedicated IPTV channel.

Inclusive usability techniques in requirements analysis of accessible web applications

To follow accessibility standards does not guarantee complete accessible web applications. There are difficulties in web application development due to not consider accessibility in software life cycle together to forget important aspects in user interaction. A proposal to evaluate the benefits of using usability techniques with inclusion in the analysis phase of a web application development is presented.

Guiding accessibility issues in the design of websites

Including accessibility in the design of Websites 1.0 entails difficulties, but this situation becomes more complicated when the user becomes a creator in the web. There are new requirements to be considered in the design due to the users active interaction in the web 2.0 environment, such as the web pages with user-generated content (like Blogs), where there are not only information receptors, but issuers as well,. This paper offers a solution for including accessibility requirements in the design process based on using accessible content templates in order to preserve accessibility and the editing process using models and rules to annotate structure and semantics of contents. This annotation is added to the final codification, which will guarantee a better complement to the standards and accessibility guidelines.

Tablet PC and Head Mounted Display for live closed captioning in education

Automatic Speech Recognition is a powerful tool for inclusive education. Captioning helps disabled students but distract other students. We propose and evaluate the use of Tablets PC and Head Mounted Displays for live closed captioning.

Web educational services for all: the APEINTA project

This paper presents the web-based educational services included in the APEINTA project. The main aim of APEINTA is to avoid barriers among the students and the education. Taking into account the advantage of cloud computing paradigm, the next web-based services are proposed: First, a captioning service, so students with hearing disabilities, for instance, could access to the verbatim speech of the teachers, even in a remote location; Second, a text-to-speech service, so students with speaking problems could participate in the class or in oral discussions or meetings, for instance, just writing in their personal devices; Third, a web-based educational system, so every student can access the pedagogical resources with time and location independency.