Current projects

Development of enabling technologies: personalization, continual dialogue understanding and explainability.

Lifelong Universal Language Representation

Learn a multilingual and multimodal model that builds upon a lifelong universal language representation.

Advanced Deep Learning Architectures for Speech, Audio and Language Processing

Tackle big challenges in multilingual and multimodal machine translation, speaker recognition, NLP and speech regeneration.

Demos and Resources

Interlingua Visualization

Web-based tool to visualize multilingual intermediate layer representations in sequence-based architectures at the sentence and token level.

MT DataSheets

An open repository for Machine Translation DataSheets. 

Past projects


Autonomous Lifelong Learning Intelligent Systems

Development of autonomous intelligent systems sustaining their performance across time.



Learn a universal language representation for multilingual text and another for multilingual speech.

Development of new architectures for speech processing with deep learning techniques.

Advanced research in core speech technologies for all the official languages of Spain.

Providing a new integrated MT paradigm which mainly includes linguistic technologies and statistical algorithms.

Development of machine translation systems which respond rapidly and intelligently to user feedback.

Research in automatic speech translation: speech recognition, statistical machine translation & speech synthesis.

Research in core language technologies for speech-to-speech translation in a multilingual context.

Development of spoken and written multilingual technologies for the design of personal assistants.

Build lexica for automatic speech recognition and text-to-speech synthesis for thirteen languages.

Provide and integrate core technologies to facilitate communication between humans.