Bangladeshi Children Language Impairment Dataset

Our dataset is focused on automatic voice recognition for the purpose of diagnosing language disorders; hence, our dataset consists of written documents. As a whole, our study dataset comprises of inquiries, assertive speech, and responses to all three. Young children are being asked these questions and taught these stories. Our focus is on children aged 0 to 6 years. We have polled numerous children within this age range with the permission of their families, including infants from our own family, extended family, neighbors, non-governmental organizations (NGOs) in Bangladesh who work with children, hospitals, and many more. The surveys focused mostly on eliciting responses from the children, whether via direct questioning or age-appropriate aggressive language. Any action or expression of approval in response to the question or statement counts as an answer. Children of varying ages reach several stages in their development of language. A baby who is 7-12 months old may utilize babbling consonant-vowel combinations and consonant sounds; a baby who is 12-20 months old may use gestures, identify their own name, etc. Prior to the experiment, we required to do an analysis of the dataset. Age, Speech in Bangla, Speech Translated in English, Response in Bangla, Response Translated in English, and label are the 6 columns that make up our dataset. We gathered information from kids as young as one month old and as old as sixty (60) months old. In addition, we can see that the number of normal samples is about twice as large as the number of impaired samples (out of 252 total samples.