For the fourth consecutive summer, ICS researcher Inés Olza has collaborated with the Google Summer of Code (GSoC). Institute for Culture and Society (ICS) Inés Olza has collaborated with the Google Summer of Code (GSoC), a global program that awards grants to young computer scientists from all over the world to collaborate with institutions, groups from research and companies dedicated to developing code for open source tools.

Inés has contributed as a mentor to the international consortium network Hen Lab for the Study of Multimodal Communication, which brings together experts from more than 20 universities in countries such as the USA, Spain, Germany, Brazil and Norway.

In this edition, he has provided advice on available text banks and linguistic tools in Spanish for applications related to automatic voice detection in this language. This is specified, for example, in the synchronization of subtitles with the image in movies and series.

The researcher explains that forced alignment is core topic to find audiovisual material in the instructions of data multimodal -those containing image, text, sound...-. She says that many of the programs that do this, such as Gentle, are highly developed in English but still not very well developed in Spanish.

From her experience as a GSoC mentor, she highlights that "it is a great opportunity to work remotely with people from other countries, learn about other disciplines, collaborate with them from linguistics and make contributions from your specialization program beyond your comfort zone".

He also stresses that Google is an example of "how the more technical knowledge can be at the service of citizens", since the code developed in this virtual summer campus is open and available to anyone Username.

12 network Hen Lab projects funded in GSoC

GSoC has funded 12 projects in 2018 from network Hen Lab. They are all related to the development of automatic text (natural language processing), sound and image processing tools that can be incorporated into its Library Services International NewsScape of Television News. This is a gigantic corpus of spoken language, making it possible to study all multimodal aspects (gesture, prosody, images and sounds accompanying speech, television production effects, etc.). This unprecedented tool could revolutionize the study of speech and news coverage.

Among other topics, the projects have addressed automatic speech recognition in various languages, with Arabic, Chinese and Russian as novelties; emotion detection and segmentation of interactions (turns, genres, conversational sequences, etc.).

In 2015, the approach of the GSoC-network Hen grants was audio analysis, while in 2016 the project focused on machine learning within the field of computer vision. The 2017 goal was to create a multimodal processing system to extract information about human communicative behavior from text, audio and video.