1 code implementation • SPECOM 2021 • Roman Vygon, Nikolay Mikhaylovskiy
In the past few years, triplet loss-based metric embeddings have become a de-facto standard for several important computer vision problems, most no-tably, person reidentification.
Ranked #1 on Keyword Spotting on Google Speech Commands
1 code implementation • 30 Mar 2021 • Rostislav Kolobov, Olga Okhapkina, Olga Omelchishina, Andrey Platunov, Roman Bedyakin, Vyacheslav Moshkin, Dmitry Menshikov, Nikolay Mikhaylovskiy
The performance of automated speech recognition (ASR) systems is well known to differ for varied application domains.
Ranked #1 on Speech Recognition on MediaSpeech
no code implementations • 24 Apr 2021 • Roman Bedyakin, Nikolay Mikhaylovskiy
This memo describes NTR-TSU submission for SIGTYP 2021 Shared Task on predicting language IDs from speech.
no code implementations • 31 May 2021 • Roman Bedyakin, Nikolay Mikhaylovskiy
In this memo, we show that a convolutional neural network with a Self-Attentive Pooling layer shows promising results in low-resource setting for the language identification task and set up a SOTA for the Low Resource ASR challenge dataset.
no code implementations • 8 Feb 2022 • Eduard Zubchuk, Dmitry Menshikov, Nikolay Mikhaylovskiy
Kiosks are a popular self-service option in many fast-food restaurants, they save time for the visitors and save labor for the fast-food chains.
no code implementations • 25 Aug 2022 • Eduard Zubchuk, Mikhail Arhipkin, Dmitry Menshikov, Aleksandr Karaush, Nikolay Mikhaylovskiy
We opensource under CC BY 4. 0 license Lib-SibGMU - a university library circulation dataset - for a wide research community, and benchmark major algorithms for recommender systems on this dataset.
no code implementations • 27 Aug 2022 • Nikolay Mikhaylovskiy
In this short note we explore what is needed for the unsupervised training of graph language models based on link grammars.
no code implementations • 11 May 2023 • Nikolay Mikhaylovskiy, Ilya Churilov
We show that the laws of autocorrelations decay in texts are closely related to applicability limits of language models.
no code implementations • 4 Jun 2023 • Nikolay Mikhaylovskiy
We propose a shared task of human-like long text generation, LTG Challenge, that asks models to output a consistent human-like long text (a Harry Potter generic audience fanfic in English), given a prompt of about 1000 tokens.
no code implementations • NAACL (SIGTYP) 2021 • Roman Bedyakin, Nikolay Mikhaylovskiy
This memo describes NTR-TSU submission for SIGTYP 2021 Shared Task on predicting language IDs from speech.