NLP

  1. Customizing Grapheme-to-Phoneme System for Non-Trivial Transcription Problems in Bangla Language publication dateFeb 2019 publication descriptionNorth American Chapter of the Association for Computational Linguistics (NAACL-HLT 2019), At Minneapolis, USA.

Scope: Grapheme to phoneme (G2P) conversion is an integral part of various text and speech processing systems, such as Text to Speech system, Speech Recognition system, etc. The existing methodologies for G2P conversion in Bangla language are mostly rule-based. However, data-driven approaches have proved their superiority over rule-based approaches for large-scale G2P conversion in other languages, such as English, German, etc. As the performance of data-driven approaches for G2P conversion depend largely on pronunciation lexicon on which the system is trained, in this paper, we investigate on developing an improved training lexicon by identifying and categorizing the critical cases in Bangla language and include those critical cases in training lexicon for developing a robust G2P conversion system in Bangla language. Additionally, we have incorporated nasal vowels in our proposed phoneme list. Our methodology outperforms other state-of-the-art approaches for G2P conversion in Bangla language.

  1. Isolated Bangla Word Recognition and Speaker Detection by Semantic Modular Time Delay Neural Network (MTDNN), pp-560-565, 18th International Conference on Computer and Information Technology (ICCIT), 21-23 December, 2015.

Scope: Developed a Deep Neural Network model for Bengali isolated acoustic words recognition over a collected corpus. Accumulated concept from Neuroscience and brain research with the application of Natural Language Processing techniques and custom Deep Neural Network. Deep Neural Architecture was designed in such a modular way that every module had some semantic role with time delay functionalities for increasing experiences over time. (Undergraduate Final Year Thesis)(Undergraduate Final Year Thesis, supervisor Moshiul Hoque, PhD.)

  1. Extracting semantic relatedness for Bangla words, Technically Co-sponsored by IEEE, IEEE CS, IEICE, IEEE TC-PAMI, IEEJ, JSME, JIN, U of Hyogo publication dateMay 2016 publication description5th International Conference on Informatics, Electronics & Vision (ICIEV)

Scope: Developed a framework for extracting semantic relational words in Bengali. Extraction of Synonyms, Antonyms, Hyponym, Hypernym, Meronym, Holonym and Polysemy are primarily investigated as a rule based model. For every word two other features: word concept(customized feature vector) and parts of speech category are also presented for clarification. Developed a semantic analyzer to extract these relations from nouns, adjectives and verbs.