Recipient of Nargis & Shahid Soorty Grant
\nan08017@st.habib.edu.pk
\nComputer Science student interested in Bioinformatics, Data Science, and Software Engineering. Passionate about research and using data-driven approaches to build impactful and meaningful technological solutions.<\/p>\n
an08017@st.habib.edu.pk
\nazkaanasir26@gmail.com
\nwww.linkedin.com\/in\/ azkaa-n-108694248
\nPortfolio<\/p>\n
+923092254119<\/p>\n
Dhanani School of Science and Engineering
\nBS Computer Science<\/p>\n
2026
\nAI|
\nData Analysis|
\nMachine Learning|
\nPython|
\nResearch|
\nSQL|
\nActs with Ownership; Adaptability; Drive for Results; Planning; Strategic Thinking; Takes Initiative
\nAI Engineer; Data Scientist; Software Engineer;
\nN\/A
\nSenior Associate Editor,Pride Press|Junior Associate Editor,Pride Press|Planning Lead,Serve Club|Operations Lead,Serve Club|
\nHBL,
\nData Science Intern
\n(June \u2013 August 2025)
\n|
\nPublication \u2013 Abstract published in Obesity Society’s Journal in November 2024|
\nYAARAN – Speech-to-Text Translator for Burushaski
\nYAARAN develops the first speech-to-text translation system for Burushaski, an endangered language isolate spoken in northern Pakistan. We created a parallel corpus of 7,115 Burushaski audio samples paired with English translations from 42 speakers, representing ~10 hours of speech data. We fine-tuned three Whisper model variants (Small: 244M, Medium: 769M, Large: 1.55B parameters) to systematically evaluate scaling benefits for low-resource translation.<\/p>\n
Using normalized evaluation metrics, we achieved BLEU 70.90 and 75% word accuracy, with Whisper-Medium offering an optimal performance-efficiency balance (BLEU 69.94). Our work establishes the first speech translation benchmark for Burushaski, demonstrates clear but diminishing returns with model scaling, and provides a replicable methodology applicable to other endangered languages worldwide.<\/p>\n","protected":false},"excerpt":{"rendered":"
Recipient of Nargis & Shahid Soorty Grant an08017@st.habib.edu.pk Computer Science student interested in Bioinformatics, Data Science, and Software Engineering. Passionate about research and using data-driven approaches to build impactful and meaningful technological solutions. an08017@st.habib.edu.pk azkaanasir26@gmail.com www.linkedin.com\/in\/ azkaa-n-108694248 Portfolio +923092254119 Dhanani School of Science and Engineering BS Computer Science 2026 AI| Data Analysis| Machine Learning| Python| […]<\/p>\n","protected":false},"author":9,"featured_media":8495,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"single-graduate-01.php","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"program":[29],"school":[42],"class":[44],"minor":[],"competencies":[57,60,61,58,63,59],"class_list":["post-8095","graduate","type-graduate","status-publish","format-standard","has-post-thumbnail","hentry","program-bscs","school-dhanani-school-of-science-and-engineering","class-class-of-2026","competencies-acts-with-ownership","competencies-adaptability","competencies-drive-for-results","competencies-planning","competencies-strategic-thinking","competencies-takes-initiative"],"acf":{"scholarship":"Recipient of Nargis & Shahid Soorty Grant","aspiration_statement":"Computer Science student interested in Bioinformatics, Data Science, and Software Engineering. Passionate about research and using data-driven approaches to build impactful and meaningful technological solutions.","video_profile_link":"","habib_email":"an08017@st.habib.edu.pk","email":"azkaanasir26@gmail.com","linkedin_link":"https:\/\/linkedin.com\/in\/ azkaa-n-108694248","portfolio_Title":"Portfolio","portfolio_link":"","personal_cell_1":"+923092254119","personal_cell_2":"","core_skills":[{"skill":"AI"},{"skill":"Data Analysis"},{"skill":"Machine Learning"},{"skill":"Python"},{"skill":"Research"},{"skill":"SQL"}],"academic_awards":[{"academic_record":"N\/A"}],"leadership-experiences":[{"leadership":"Senior Associate Editor, Pride Press"},{"leadership":"Junior Associate Editor, Pride Press"},{"leadership":"Planning Lead, Serve Club"},{"leadership":"Operations Lead, Serve Club"}],"work-experience":[{"experience":"HBL, Data Science Intern (June \u2013 August 2025)"}],"publications-projects":[{"publications_title":"","publications":"Publication \u2013 Abstract published in Obesity Society's Journal in November 2024"}],"project_title":"YAARAN - Speech-to-Text Translator for Burushaski","fyp_desc":"YAARAN develops the first speech-to-text translation system for Burushaski, an endangered language isolate spoken in northern Pakistan. We created a parallel corpus of 7,115 Burushaski audio samples paired with English translations from 42 speakers, representing ~10 hours of speech data. We fine-tuned three Whisper model variants (Small: 244M, Medium: 769M, Large: 1.55B parameters) to systematically evaluate scaling benefits for low-resource translation.\r\n\r\nUsing normalized evaluation metrics, we achieved BLEU 70.90 and 75% word accuracy, with Whisper-Medium offering an optimal performance-efficiency balance (BLEU 69.94). Our work establishes the first speech translation benchmark for Burushaski, demonstrates clear but diminishing returns with model scaling, and provides a replicable methodology applicable to other endangered languages worldwide.","project_pictures":[{"project_pic":8877}],"project-video":""},"yoast_head":"\n