INESC-ID

About: INESC-ID is a nonprofit organization based out in Lisbon, Portugal. It is known for research contribution in the topics: Field-programmable gate array & Control theory. The organization has 932 authors who have published 2618 publications receiving 37658 citations.

The Titech large vocabulary WFST speech recognition system

Abstract: In this paper we present evaluations on the large vocabulary speech decoder we are currently developing at Tokyo Institute of Technology. Our goal is to build a fast, scalable, flexible decoder to operate on weighted finite state transducer (WFST) search spaces. Even though the development of the decoder is still in its infancy we have already implemented a impressive feature set and are achieving good accuracy and speed on a large vocabulary spontaneous speech task. We have developed a technique to allow parts of the decoder to be run on the graphics processor, this can lead to a very significant speed up.

Grid code reinforcements for deeper renewable generation in insular energy systems

Abstract: Introduction of renewable energy sources (RES) in insular areas is growing on different islands of various regions in the world and the large-scale deployment of renewables in island power systems is appealing to local attention of grid operators as a method to decrease fossil fuel consumption. Planning a grid based on renewable power plants (RPP) presents serious challenges to the normal operation of a power system, precisely on voltage and frequency stability. Despite of its inherent problems, there is a consensus that in near future the RES could supply most of local needs without depending exclusively on fossil fuels. In previous grid code compliance, wind turbines did not required services to support grid operation. Thus, in order to shift to large-scale integration of renewables, the insular grid code ought to incorporate a new set of requirements with the intention of regulating the inclusion of these services. Hence, this paper discusses grid code requirements for large-scale integration of renewables in an island context, as a new contribution to earlier studies. The current trends on grid code formulation, towards an improved integration of distributed renewable resources in island power systems, are addressed. The paper also discusses advanced grid code requirement concepts such as virtual wind inertia and synthetic inertia for improving regulation capability of wind farms and the application of energy storage systems (EES) for enhancing renewable generation integration. Finally, a comparative analysis of insular grid code compliance to these requirements is presented in the European context.

Using Machine Learning to Improve the Prediction of Functional Outcome in Ischemic Stroke Patients

Abstract: Ischemic stroke is a leading cause of disability and death worldwide among adults. The individual prognosis after stroke is extremely dependent on treatment decisions physicians take during the acute phase. In the last five years, several scores such as the ASTRAL, DRAGON, and THRIVE have been proposed as tools to help physicians predict the patient functional outcome after a stroke. These scores are rule-based classifiers that use features available when the patient is admitted to the emergency room. In this paper, we apply machine learning techniques to the problem of predicting the functional outcome of ischemic stroke patients, three months after admission. We show that a pure machine learning approach achieves only a marginally superior Area Under the ROC Curve (AUC) ( $0.808\pm 0.085$ ) than that of the best score ( $0.771\pm 0.056$ ) when using the features available at admission. However, we observed that by progressively adding features available at further points in time, we can significantly increase the AUC to a value above 0.90. We conclude that the results obtained validate the use of the scores at the time of admission, but also point to the importance of using more features, which require more advanced methods, when possible.

Age and gender classification using fusion of acoustic and prosodic features.

Abstract: This paper presents a description of the INESC-ID Spoken Language Systems Laboratory (L2F) Age and Gender classification system submitted to the INTERSPEECH 2010 Paralinguistic Challenge. The L2F Age classification system and the Gender classification system are composed respectively by the fusion of four and six individual sub-systems trained with short and long term acoustic and prosodic features, different classification strategies (GMM-UBM, MLP and SVM) and using four different speech corpora. The best results obtained by the calibration and linear logistic regression fusion back-end show an absolute improvement of 4.1% on the unweighted accuracy value for the Age and 5.8% for the Gender when compared to the competition baseline systems in the development set.