Research areas

“Machine Learning is the new electricity” Deep Learning is a technique that belongs to the Machine Learning Field. Machine Learning techniques learns from data. Nowadays the amount of data grows exponentially year after year. Therefore machine learning techniques obtain a great potential to solve very complex problems. Big-data is the perfect partner and deep learning techniques are becoming a standard thanks [...]

Read more

Speech-to-speech translation or text-to-text translation for limited domains fall within these kind of projects. Finite-state and statistical transducers are used as the basis of the machine translation systems. These models can be learnt automatically from real examples of translation. Some applications included (but are not limited to) translation of technical reports, hotel services, Speech interaction with mobile devices Speaker and domain adaptation Statistical [...]

Read more

Both off-line (document images) and on-line HTR (tablet or e-pen signals) are considered. No prior character or word segmentation is needed. Technology, borrowed from Speech Recognition, relies on character Hidden Markov Models, Finite State word models, and syntactic N-Grams. After model training, for each given text line image, a holistic (“Viterbi”) search provides both an optimal transcription and the corresponding [...]

Read more

General Statistical and Syntactic Pattern Recognition techniques for image analysis and recognition. Some applications: OCR and document analysis, medical diagnosis, biometric identification, image and video retrieval. Relevance-based Image Retrieval Biometrics

Read more

The activities of the Machine Translation group began some years ago with the use of finite-state models for speech-to-speech translation and for text-to-text translation in limited domains. This group has developped a number of translation models with the corresponding learning algorithms and a number of prototypes for speech translation and computer-assisted translation. Currently, the Machine Translation group is devoted to the [...]

Read more

For many languages that use non-Roman based indigenous scripts (e.g., Arabic, Greek and Indic languages) one can often find a large amount of user generated transliterated content on the Web in the Roman script. IR in such space is challenging because queries written in either the native or the Roman scripts need to be matched to the documents written in [...]

Read more