SpeechTech4All is a project focused on advanced research in all core speech technologies (speech recognition, automatic translation and text to speech conversion) for all the official languages of Spain. It is also aimed at the recognition of the speaker’s emotional state and the building of multimodal (speech and face) and multilingual (Spanish, Galician, Catalan and Basque) experimental work-frames that allow showing the work developed.
As a result of the project significant research advances will be obtained in each of the above mentioned technologies. Some of these advances are: universal access to the service of synthetic voices customization, development of domain adaptation techniques in automatic translation and development of emotional state detection systems based in speech and face joint processing. The project will take part in the international competitive evaluation campaigns organized by Interspeech and the Red Temática de Tecnologías del Habla (RTTH).
To show the advances achieved in all the technologies considered, as well as to manifest the social character of the project, two showcases are defined:
1) The first one integrates most of the technologies considered in the project and consists on the multilingual subtitling and dubbing of audiovisual material related to education: documentaries, speeches, seminars…
2) The second one is focused on one of the true applications of speech technologies: providing speech to people who have a severe level of oral disability by the use of a speech synthesis system adapted to the specific characteristics of the speaker.PARTNER: Universitat Politècnica de Catalunya
FUNDING INSTITUTION: Ministerio de Economía y Competitividad
DURATION: 2012-2015
Linguistic diversity is an essential element of our mulitcultural European society. In the age of globalisation and the emerging information society this diversity is not yet sustainable. Languages that are not supported by advanced information technology will be in danger. The vision of an inclusive barrier-free European information society requires technologies for overcoming language boundaries at affordable costs. The creation of the technology base for the multilingual European information society needs a major joint effort of research, industrial technology providers, commercial and public user organisations, policy makers and research funding agencies. T4Me is a new Network of Excellence founded by thirteen European research and technology organizations. Its mission is to forge a strategic alliance for creating the technologies and applications needed for making the linguistic diversity of our multicultural society sustainable.
PARTNER: Fundació Barcelona Media
FUNDING INSTITUTION: European Comission
DURATION: 2010-2013
- BUCEADOR – Information search engine for multilingual audiovisual contents
BUCEADOR is a project focused on advanced research in all core Spoken Language Technologies (SLT), (diarization, speech recognition, speech machine translation, and text-to-speech conversion), research in both, cross-language and speech-oriented information retrieval technologies for voice search applications, and the successful joint integration of all of them in a multilingual and multimodal information retrieval system.The goal of the project is to achieve improvements in all the SLT components and voice search applications to improve human-machine and human-to-human communication among all the official languages spoken in Spain as well as between these languages and English.
PARTNER: Fundació Barcelona Media / Universitat Politècnica de Catalunya
FUNDING INSTITUTION: Ministerio de Economía y Competitividad
DURATION: 2010-2012
Searching for information in long videos can be a time-consuming experience. In order to deal with this problem, we have developed the OnAIR, an ontology-aided information retrieval system, which aims to retrieve clips from video collections. It allows users to enter a query in natural language and searches a digital video database for the clip which best answers the query. Its implementation uses a domain ontology to improve the quality of the retrieved information. The power of description logics and the reasoning mechanisms make it possible to infer new relationships between concepts (at the domain ontology) that would, otherwise, remain unknown. The system uses such information during the search for videos, at a process called query expansion, where the new related concepts are used to enrich the original query. We have tested the system with a series of queries for the Contemporary Art case study and verified that the use of these semantic web techniques resulted in a relevance improvement of the retrieved documents. We have designed the system to work in a domain-independent way, allowing us to move to other domains by just changing the underlying domain ontology and video collection. OnAIR is maintained and developed by a team of graduate and undergraduate students at IME-USP. We are currently working on a Web 2.0 interface and ways to improve the domain ontology building task.
PARTNER: Universidade de Sao Paulo
FUNDING INSTITUTION: FAPESP
DURATION: 2004-2012
- Cénit Integra: Investigación en Tecnologías para la Gestión de la Migración
El objetivo global del proyecto es investigar y desarrollar tecnologías que permitan realizar la prevención, el control y la integración de gestión de la inmigración, cubriendo todo el espacio fronterizo.
PARTNER: Fundació Barcelona Media
FUNDING INSTITUTION: Ministerio de Economía y Competitividad
DURATION: 2008-2011
The main objective of i3media is to generate a framework for the representation and manipulation of intelligent digital contents and for the development of technologies which enable a higher automatization of the audio-visual production. In particular, the increase with respect to the state of the art will involve the automatic association of the intelligence and auto-adaptation functions with the content itself, independently of the underlying communication networks.
PARTNER: Fundació Barcelona Media
FUNDING INSTITUTION: Ministerio de Economía y Competitividad
DURATION: 2007-2010
- Avivavoz. Voice translation technologies: recognition, statistical machine translation based on corpus and synthesis.
PARTNER: Universitat Politècnica de Catalunya
FUNDING INSTITUTION: Ministerio de Educación y Ciencia
DURATION: 2007-2009
PARTNER: Universitat Politècnica de Catalunya
FUNDING INSTITUTION: Generalitat de Catalunya
DURATION: 2007-2009
- TC-STAR: Technology and Corpora for Speech to Speech translation
PARTNER: Universitat Politècnica de Catalunya
FUNDING INSTITUTION: European Comission
DURATION: 2004-2007
- ALIADO: Tecnologías del habla y el lenguaje para un asistente personal
PARTNER: Universitat Politècnica de Catalunya
FUNDING INSTITUTION: Ministerio de Educación
DURATION: 2004-2006
Créer une plate-forme d’indexation de l’audio afin de permettre à des utilisateurs de retrouver une information dans un corpus d’émissions radiophoniques de Radio France comme s’il s’agissait de documents texte.
PARTNER: LIMSI-CNRS
FUNDING INSTITUTION: Résau National en Technologies Logicielles
DURATION: 2002-2004