Aditya Yadavalli
Speech/NLP Engineer at Karya
I am a Speech/NLP Engineer at Karya. Here, I work on data processing pipelines, quality estimation of crowdsourced datasets, and building assistive tools for crowdsource workers.
I am also a visiting researcher at Case Western Reserve University (CWRU), where I collaborate with Prof. Vera Tobin. Here, I work on evaluating NLP models trained on Child Directed Speech (CDS) and am looking to establish common mistakes that humans and NLP models make when acquiring a new language.
In my free time, I volunteer at Masakhane – a group dedicated to work on improving NLP models for Africa. Along with the rest in the group, I explore how current NLP models do not generalise well, especially when we encounter African languages and accents, and how we can make them better.
Previously, I was a dual degree student at IIIT Hyderabad, where I pursued B.Tech. (Hons) in Computer Science & M.S. by Research in Computational Linguistics. For my M.S. thesis, I explored how closely related languages can be used to improve the performance of low-resource languages at Speech Processing Lab (SPL) with Prof. Anil Kumar Vuppala.
When not at work, you can find me discussing cricket or picking up obscure trivia that no one cares about.
news
Mar 28, 2024 | Paper titled “Speaking in Terms of Money: Financial Knowledge Acquisition through Speech Data Generation” accepted at COMPASS 2024! |
---|---|
Mar 25, 2024 | Paper titled “Akal Badi ya Bias”: An Exploration Study of Gender Bias in Hindi accepted at FAccT 2024! |
Jan 27, 2024 | Paper titled MunTTS : A Text-to-Speech System for Mundari accepted at ComputEL-7 workshop |
Jan 27, 2024 | Paper titled AccentFold: A Journey through African Accents for Zero-Shot ASR Adaptation to Target Accents accepted at EACL 2024 Findings |
Oct 4, 2023 | Paper titled AfriSpeech-200: Pan-African Accented Speech Dataset for Clinical and General Domain ASR accepted in TACL & will be presented at EMNLP 2023! |
May 4, 2023 | Paper titled SLABERT Talk Pretty One Day: Modeling Second Language Acquisition with BERT accepted at ACL 2023 ![]() |
May 4, 2023 | Paper titled X-RiSAWOZ: High-Quality End-to-End Multilingual Dialogue Datasets and Few-shot Agents accepted in ACL Findings! |
Nov 11, 2022 | Defended my MS thesis! Thanks to my panel – Dr. Kishore Prahallad and Prof. Chiranjeevi Yarra ![]() |
Oct 3, 2022 | Paper on “How Do Phonological Properties Affect Bilingual Automatic Speech Recognition?” accepted at IEEE SLT 2022 ![]() |
Sep 10, 2022 | I will visiting Incheon to attend Interspeech! Happy to meet you if you’ll be attending the same. |
Jul 1, 2022 | I will be attending NAACL in person. If you are attending too, let’s catch up! |
Jun 19, 2022 | Submitted my MS Thesis for review ![]() |
Jun 15, 2022 | Paper on “Multi-Task End-to-End Model for Telugu Dialect and Speech Recognition” accepted in Interspeech 2022 ![]() |
Jun 1, 2022 | Started working at Karya under the mentorship of Dr. Vivek Seshadri as a Speech/NLP Engineer. |