Learning to Recognize Dialect Features

Building NLP systems that serve everyone requires accounting for dialect differences. But dialects are not monolithic entities: rather, distinctions between and within dialects are captured by the presence, absence, and frequency of dozens of dialect features in speech and text, such as the deletion of the copula in He ∅ running''. In this paper, we introduce the task of dialect feature detection, and present two multitask learning approaches, both based on pretrained transformers. For most dialects, large-scale annotated corpora for these features are unavailable, making it difficult to train recognizers. We train our models on a small number of minimal pairs, building on how linguists typically define dialect features. Evaluation on a test set of 22 dialect features of Indian English demonstrates that these models learn to recognize many features with high accuracy, and that a few minimal pairs can be as effective for training as thousands of labeled examples. We also demonstrate the downstream applicability of dialect feature detection both as a measure of dialect density and as a dialect classifier.

" The Third Language 'in-between' in the Novels of Ahmed Youssef Dawood"

Language is one means of communication that has the most significant role in enhancing humans' life and their relation with their environment alongside their relations with the society in which they were born and raised. Language has always been the product of this society on whose progress and regress have an impact upon it. It is well-known that standard Arabic is the official language with its accurate grammar and vocabulary moving from the ancestor to the descendant. However, it very often may be difficult to apply or have access to for most people regardless of their cultural qualifications. It is also difficult for this language to convey or transfer reality as clear as it is or to express how easy and spontaneous life is to all people. Since the phenomenon of vernacular language alongside standard language is a linguistic one all over the world, thus the necessity in the Arabic novel in general and countryside in particular emerged to have an in-between third language that is neither standard nor vernacular. This novel language is to be capable of bringing the standard closer to daily life and ending up with one form of dialogue that provides characters with their psychological and social traits; a tacit language for all different cultural and scientific levels of readers and their social status. Also, this language will help the text express the human emotions that emerge subconsciously for the standard one is incapable of doing so. Needless to say, standard Arabic was one day a vernacular with different dialects expressed through words like "language" and "tongue." Allah said: ("We have not sent but a messenger to represent his nation and clarify the truth to them. For, God guide and misguide whomsoever thus He is the Noble and Wise").

The approach of the ancients in dealing with dialects and the position of modernists

The subject of dialects in Arabic grammar is a subject of confusion and confusion at the time of Arabic grammar when they used the term "dialect" and "language" in their expression of the dialectic differences between the tribes. The modernists, moreover, did not have independent works, which were specialized in studying each dialect separately. They identified the clear tribes that could be adopted in their language, and left the other tribes under the pretext of leaving the linguistic level. Therefore, my dependence on this research will focus on two issues: The ancients in their dealings with dialects, relying on what was in the book properties of Ibn-taking, book Sibawayh, book Alsahabay in the jurisprudence of the language of IbnFaris, a book brief cooler and other books on this subject, and the second: modern attitude and the most prominent criticism of the approach to the ancients.