In concatenative languages such as English, the morphemes of a word are linked linearly so that words formed from the same base morpheme also resemble each other along orthographic dimensions. In Hebrew, by contrast, the morphemes of a word can be but are not generally concatenated. Instead, a pattern of vowels is infixed between the consonants of the root morpheme. Consequently, the shared por...

Early morphological decomposition of complex words has been supported by evidence showing that the magnitude of masked transposed-letter (TL) priming effects is greater for within-morpheme transpositions than for between-morpheme transpositions. However, these findings have lately been called into question, and a recent article by Sánchez-Gutiérrez and Rastle (Psychonomic Bulletin & Review, 20,...

Traditional word embedding models only learn word-level semantic information from corpus while neglect the valuable semantic information of words’ internal structures such as morphemes. To address this problem, the goal of this paper is to exploit the morphological information to enhance the quality of word embeddings. Based on spectral method, we propose two word embedding models: Morpheme on ...

This paper describes development of a large vocabulary continuous speaker independent speech recognition system for Estonian. Estonian is an agglutinative language and the number of different word forms is very large, in addition, the word order is relatively unconstrained. To achieve a good language coverage, we use pseudo-morphemes as basic units in a statistical trigram language model. To im...

This paper describes speech translation from Amharic-to-English, particularly Automatic Speech Recognition (ASR) with post-editing feature and AmharicEnglish Statistical Machine Translation (SMT). ASR experiment is conducted using morpheme language model (LM) and phoneme acoustic model (AM). Likewise, SMT conducted using word and morpheme as unit. Morpheme based translation shows a 6.29 BLEU sc...

Morpho Challenge is an annual evaluation campaign for unsupervised morpheme analysis. In morpheme analysis, words are segmented into smaller meaningful units. This is an essential part in processing complex word forms in many large-scale natural language processing applications, such as speech recognition, information retrieval, and machine translation. The discovery of morphemes is particularl...

Palatalization in Korean is of two types lexical palatalization governed by language-specific phonological rules, and postlexical palatalization that appears to be purely phonetic. While lexical palatalization only occurs when a morpheme boundary intervenes between a target consonant and a palatalization trigger, post-lexical palatalization occurs irrespective of the presence of a morpheme boun...

We extend the unsupervised morpheme segmentation method Morfessor Baseline to account for the linguistic phenomenon of allomorphy, where one morpheme has several different surface forms. Our method discovers common base forms for allomorphs from an unannotated corpus. We evaluate the method by participating in the Morpho Challenge 2008 competition 1, where inferred analyses are compared against...

The present study explores the effects of frequency in learning to parse novel morphological patterns. In two experiments, suffixes were divided into three classes: high, medium and low frequency, based on the proportion of stems in the input that each suffix attached to (high frequency = 12/12, medium frequency = 6/12, and low frequency = 2/12). In Experiment 1, learners were better at segment...

In this paper, we present an unsupervised algorithm for morpheme discovery called UNGRADE (UNsupervised GRAph DEcomposition). UNGRADE works in three steps and can be applied to languages whose words have the structure prefixes-stem-suffixes. In the first step, a stem is obtained for each word using a sliding window, such that the description length of the window is minimised. In the next step p...