— Strong knowledge of digital audio signal processing theory and principles;
— Experience with speech recognition software;
— Knowledge of Digital Signal Processing theory;
— Experience with machine learning framework PyTorch, TensorFLow;
— Knowledge of CUDA / interest in developing custom CUDA kernels;
— Eager Ability to follow the state-of-the art learn new knowledge in speech related technologies.
Will be a plus
— Experience with implementing algorithms in C/C++;
— Knowledge of Voice Activity Detection algorithms.
— Do real research with models of audio in a small, scientifically-minded team;
— Freedom to try out ideas of your own;
— Challenging work on an interesting project;
— Very warm and friendly working environment;
— Flexible working hours
— Research of speech recognition algorithms;
— Implementation of signal processing tasks;
— Text Crawling, Processing, and Normalization;
— Writing unit tests for algorithmic projects.