The Main Characteristics Of Amharic Language
Amharic is one of the Ethio-Semitic languages, which belongs to the Semitic branch of the Afroasiatic super family and is related to Hebrew, Arabic, and Syrian. Amharic, which is spoken mainly in Ethiopia, is the second most spoken Semitic language, after Arabic. It is spoken by several million people as a first language and as a second language throughout different regions of Ethiopia. Currently, Amharic is the official working language of the federal democratic republic of Ethiopia.
Amharic has mainly five dialectical variations (Addis Ababa, north shoa, Wollo, Gojjam, and Gondar) spoken in different regions of the country. The speech of Addis Ababa has emerged as the standard dialect and has wide spread across all Amharic-speaking communities.
The main characteristic of Amharic is the complex morphology where many new words can be derived from a single stem by addition of several prefixes, infixes and suffixes. Although there is not a one to one correspondence between Amharic morphemes and English words, we can say that one Amharic word may correspond to a group of English words. This complexity nature causes the vocabulary to expand significantly which is problematic for speech recognition.
Another characteristic of Amharic is the free words order where the order of constituents can be changed without affecting the grammaticality of a sentence. In terms of constituent orders, Amharic can be considered as a Subject-Object-Verb (SOV) type language, however, other constituent orders are also common. Free word order causes data sparseness which leads to non-robust language Model (LM) estimates. As with all of the other languages, Amharic has its own characterizing phonetic, phonological and morphological properties. It has a set of speech sounds that is not found in other languages. For example the following sounds are not found in English: ጸ, ፀ, ኸ, ቀ, ጨ, etc. Amharic speech contains 38 different phones with 31 consonants and 7 vocal sounds (vowels) and at least 234 distinct CV (consonant-vowel) syllables. The consonants are generally classified as stops, fricatives, nasals, liquids, and semi-vowels. Amharic spoken document retrieval is an important problem because Amharic is one of the key languages used by majority of population of Ethiopia, and much information exist in the form of Amharic spoken audio. The linguistic properties of Amharic speech are very different from that of English. This affects the techniques used in Amharic spoken document retrieval. Amharic suffers from data sparseness and out of vocabulary words problems as it is a morphologically rich language. Therefore, the development of subword based language models for Amharic is recommended.
