首页>
外国专利>
Audiotextuales automatic indexing of documents based on their difficulty of understanding.
Audiotextuales automatic indexing of documents based on their difficulty of understanding.
展开▼
机译:音频文本根据其理解的难易程度自动为文档建立索引。
展开▼
页面导航
摘要
著录项
相似文献
摘要
Method for indexing a digital audiotextual document acquired in a database (SGBD) of a device (SD) digital processing of documents and comprising a signal (ST) text and by a signal (SA) synchronized audio, wherein It includes an analysis (E2) portions textual (PT1 to PTN) extracted from the text signal primarily in order to segment portions into words and marking the portions by respective time boundaries (ti, tf), an evaluation (E51) of number (nms) of syllables in words of each portion analyzed (PTn), an evaluation (E52) of the speed (ven) of elocution of each portion analyzed equal to the ratio between the number of syllables and the duration of the portion between the time boundaries (ti, tf) thereof in order to evaluate (E54) the speed (VE) of elocution of the document (DAT) according to the evaluated speeds of elocution of specified analyzed portions, a determination (E 8) of an index (IDC) comprehensibility difficulty document based on at least a first index (IVE) resulting from a comparison (E55) of the speed of elocution of the document evaluated with at least a first threshold (UV1; UV2), and a formation (E4) of a label of the document with at least the first index (IVE) and difficulty index (IDC) of understanding in a database (SGBD) of the treatment device document (SD) .
展开▼