This paper describes a pilot study in which we collected a small learner corpus of Arabic, developed a tagset for errorannotation of Arabic learner data, tagged the data for error, and performed simple Computer-aided Error Analysis (CEA). Language Learner Corpora and Applications Learner corpora research uses the methods and tools of Second Language Acquisition (SLA) studies and corpus linguist...

Pronunciation scoring is one important task for software designed to give feedback to students practicing a second language. English intonation can convey information about a speaker’s nativeness, so previous studies have proposed using intonation-based models to score nonnative pronunciation. One past approach trained models for a set of pronunciation scores using ad hoc features derived from ...

Tests of foreign accent usually treat native listeners as reliable providers of accentedness ratings, and pay too little heed to task-specific effects on non-native speakers' performance. This paper details a number of factors which in fact influence native listeners' perceptions, and the nativelike behaviour of non-native speakers' productions, based on the results of a large study of phonetic...

•The frequency of prosodic events has an impact on the perception of nativeness and fluency: – Liscombe (2007): distances between high boundary tones correlates with higher pronunciation scores. – Rosenberg (2009): higher rate of pitch accenting for of Mandarin Chinese reading English segments These stuides rely on ToBI annotations. How do these labels apply to non-native speech? •Native/non-na...

The topic of this paper is the analysis a series nine texts by Ivo Kalinski dedicated to Dragutin Domjanić Festivity Kajkavian Children's Poetry, which held in Sveti Ivan Zelina. articles were written each year from 1992 2003 and published book “Anatomija kmice ili Umjetnina teksta – zanos i tjeskoba“ (2004). These are poetological makes various observations about children's poetic texts, his s...

In this paper, we explore the differences between direct and symbolic sequential modeling of prosody. We use sequential models to characterize speech in two tasks, classifying speaking-style and distinguishing native from non-native speech. We explore the use of a spike-and-slab model to directly model pitch contour data. We find in both of these tasks that sequences of symbolic prosodic events...