Tatiana Likhomanenko

Tatiana is a postdoc researcher with Facebook's speech team, working on semi-supervised and unsupervised approaches for speech recognition systems. She holds an PhD in computer science from Lomonosov Moscow State University. Prior to joining Facebook, she worked on applications of machine learning to high-energy physics as a researcher in the scientific joint lab at Yandex, a Russian search-engine company, and at NTechLab, a leader in face recognition.

Tatiana's Publications

April 07, 2020

RESEARCH

SPEECH & AUDIO

Who Needs Words? Lexicon-Free Speech Recognition

Lexicon-free speech recognition naturally deals with the problem of out-of-vocabulary (OOV) words. In this paper, we show that character-based language models (LM) can perform as well as word-based LMs for speech recognition, in word error…

Tatiana Likhomanenko, Gabriel Synnaeve, Ronan Collobert,

April 07, 2020

April 07, 2020

RESEARCH

NLP

Scaling up online speech recognition using ConvNets

We design an online end-to-end speech recognition system based on Time-Depth Separable (TDS) convolutions and Connectionist Temporal Classification (CTC). The system has almost three times the throughput of a well tuned hybrid ASR baseline…

Vineel Pratap, Qiantong Xu, Jacob Kahn, Gilad Avidov, Tatiana Likhomanenko, Awni Hannun, Vitaliy Liptchinsky, Gabriel Synnaeve, Ronan Collobert,

April 07, 2020