Aleksandr Laptev

Aleksandr Laptev is a Ph.D. student at ITMO University and a senior research scientist at NVIDIA. His scientific interests are automatic speech recognition, speech synthesis (TTS), and natural language processing. He writes open-access scientific articles, contributes to open-source software, and participates in international speech recognition competitions. His current research area is differentiable weighted finite-state transducers.
Avatar photo

Posts by Aleksandr Laptev

Person standing in front of waterfall
Conversational AI

Entropy-Based Methods for Word-Level ASR Confidence Estimation

Once you have your automatic speech recognition (ASR) model predictions, you may also want to know how likely those predictions are to be correct. This... 12 MIN READ
Decorative image.
Conversational AI

Changing CTC Rules to Reduce Memory Consumption in Training and Decoding

Loss functions for training automatic speech recognition (ASR) models are not set in stone. The older rules of loss functions are not necessarily optimal.... 8 MIN READ