The goals of the IPSOM project are the indexing, integration and retrieval of sound in multimedia documents, applied to spoken books. With this project, we intend to broaden and improve the use of the collection of spoken books of Biblioteca Nacional (The National Library). On one hand, making them available in digital format will simplify their access to a potentially much broader community of users. On the other hand, through the alignment between the written and spoken representation, allowing the indexing and retrieval of words, sets of words or topics, the project will open new horizons for the use of these documents.
The results of this project will be particularly relevant for visually impaired citizens, allowing a non-sequential search and access to the information stored in spoken books. The multimedia applications to be developed will allow the access to a very large spoken corpus for didactic or other purposes.
LASIGE is supported by FCT project with reference UID/CEC/00408/2013.