Aligning and recognizing spoken books in different varieties of Portuguese

Spoken language technologies applied to digital talking books

“…Recently, and in the scope of a cooperation agreement with the University of Rio Grande do Sul, we have extended this repository with books read in Brazilian Portuguese (BP). The availability of the same book in different varieties of Portuguese has already showed that DTBs can be useful for research on the differences between varieties of the same language [2]. This study is currently being extended to the varieties spoken in African countries with Portuguese as the official language.…”

“…In order to mimic the interaction possibilities of print books, the standard defines a minimum set of features for a computerbased DTB player: no need to use visual display to operate device; variable playback speed; document accessible at fine level of detail; usable table of contents; easy skips (moving sequentially through the elements); ability to move directly to a specific target; setting and labeling bookmarks; ability to add information (highlighting and notes); presentation of visual elements in alternative formats (speech); etc. This standard is also known as DAISY 3.0, a designation that reflects the major work done in the area by the Daisy Consortium 2 . The standard focuses on the files, their structure and content, but does not include specifications for playback devices.…”