About

Shreekantha Nadig is an accomplished Speech and Language Technology Engineer with over 6 years of experience in both industry and academic research labs. He holds an MS by Research in Speech Recognition from the International Institute of Information Technology, Bangalore. Currently, he is working as a Speech Recognition Engineer at Dialpad, where he is developing the next-gen ASR system. At Dialpad, Shreekantha has architected and built a next-gen Speech Recognition product end-to-end from R&D to production that is toolkit agnostic and performing better than HMM hybrid ASR models. He has also led the R&D on streaming end-to-end ASR for conversational, telephony, and videoconferencing speech under low latency and multi-accent scenarios. Shreekantha has benchmarked various toolkits, including Kaldi, K2, ESPnet, NeMo, and WeNet to architect the next-gen ASR system. Additionally, he has trained and benchmarked various end-to-end ASR architectures with CTC, Attention-based Encoder-Decoder (AED), Transducer, Transformer, and Conformer models with hybrid ASR models and external ASR services. Shreekantha has developed interfaces for the shallow fusion of multi-level (sub-word and word) RNNLMs and n-gram LMs and methods to bias the models towards a list of keywords, resulting in an absolute WERR of 7%. He has also automated the data preparation pipeline for training ASR models, reducing the turnaround time for experiments and increasing the productivity of the team. In the past, Shreekantha worked as a Machine Learning Intern - ASR at ObserveAI, where he developed a feature extraction pipeline using tf.signal and tf.data, implemented different keyword-spotting (KWS) papers - Deep-KWS, CTC KWS, and deployed the KWS model using TensorFlow serving with an RTF of 0.05 on GPU. He also worked as a Graduate Teaching Assistant at the International Institute of Information Technology – Bangalore, where he taught Deep learning for Automatic Speech Recognition. Shreekantha has a strong educational background, including a Master of Science by Research in Speech Recognition from the International Institute of Information Technology – Bangalore, a Bachelor of Engineering in Telecommunications Engineering from J N N College of Engineering, Shimoga, and a Pre-University degree in Computer Science from Sri Vidyabharathi Pre-University College, Shimoga. He