Topic segmentation traditionally relies on lexical cohesion measured through word re-occurrences to output a dense segmentation, either linear or hierarchical. In this paper, a novel organization of the topical structure of textual content is proposed. Rather than searching for topic shifts to yield dense segmentation, we propose an algorithm to extract topically focused fragments organized in ...

Purely statistical models have accounted for infants' early ability to segment words out of fluent speech, with Bayesian models performing best (Goldwater et al. 2009). Yet these models often incorporate unlikely assumptions, such as infants having unlimited processing and memory resources and knowing the full inventory of phonemes in their native language. Following Pearl, et al. (2011), we ex...

The main goal of our current research is the development of the Swedish prosody model. In our analysis of discourse and dialogue intonation we are exploiting model-based resynthesis. By comparing synthesized default and fine-tuned pitch contours for dialogues under study we are able to isolate relevant intonation patterns. This analysis of intonation is related to an independent modelling of to...

Laughter is a key element of human-human interaction, occurring surprisingly frequently in multi-party conversation. In meetings, laughter accounts for almost 10% of vocalization effort by time, and is known to be relevant for topic segmentation and the automatic characterization of affect. We present a system for the detection of laughter, and its attribution to specific participants, which re...

739 c ï¿¿ ATALA

This paper describes our contribution to the English Entrance Exams task of CLEF 2015, which requires participating systems to automatically solve multiple choice reading comprehension tasks. We use a combination of text segmentation and different similarity measures with the aim of exploiting two observed aspects of tests: 1) the often linear relationship between reading text and test question...

Similarity in contextual behavior between words is considered a source of “lexical cohesion,” which is otherwise hard to measure or quantify. Such contextual similarity is used by an implementation for text segmentation, the VecTile system, which uses precompiled vector representations of words to produce similarity curves over texts. The performance of this system is shown to improve over that...

There is abundance of digitised texts available in Sanskrit. However, the word segmentation task in such texts are challenging due to the issue of Sandhi. In Sandhi, words in a sentence often fuse together to form a single chunk of text, where the word delimiter vanishes and sounds at the word boundaries undergo transformations, which is also reflected in the written text. Here, we propose an a...

In English written text, words are separated by spaces, but in written Chinese text, there are no such separators between words. (See Figure 1.) Thus, effective information retrieval of Chinese text first requires good word segmentation. In this paper, we investigate an efficient algorithm to discover the words and their occurrence probabilities from a corpus of unsegmented text without using a...