Zobrazeno 1 - 10
of 360
pro vyhledávání: '"P, Girin"'
Autor:
Girin, H., Bittner, S., Checoury, X., Decanini, D., Dietz, B., Grigis, A., Lafargue, C., Zyss, J., Xu, C., Sebbah, P., Lebental, M.
Classical and wave properties of microlasers with the shape of a truncated pseudosphere are investigated through experiments and numerical simulations. These pseudosphere microlasers are surface-like organic microlasers with constant negative curvatu
Externí odkaz:
http://arxiv.org/abs/2410.07034
Most speech self-supervised learning (SSL) models are trained with a pretext task which consists in predicting missing parts of the input signal, either future segments (causal prediction) or segments masked anywhere within the input (non-causal pred
Externí odkaz:
http://arxiv.org/abs/2405.20101
In this paper, we propose a latent-variable generative model called mixture of dynamical variational autoencoders (MixDVAE) to model the dynamics of a system composed of multiple moving sources. A DVAE model is pre-trained on a single-source dataset
Externí odkaz:
http://arxiv.org/abs/2312.04167
This work builds on a previous work on unsupervised speech enhancement using a dynamical variational autoencoder (DVAE) as the clean speech model and non-negative matrix factorization (NMF) as the noise model. We propose to replace the NMF noise mode
Externí odkaz:
http://arxiv.org/abs/2306.07820
In this paper, we present a multimodal and dynamical VAE (MDVAE) applied to unsupervised audio-visual speech representation learning. The latent space is structured to dissociate the latent dynamical factors that are shared between the modalities fro
Externí odkaz:
http://arxiv.org/abs/2305.03582
Autor:
A., Girin I., F., Likhachev S., S., Andrianov A., S., Burgin M., V., Popov M., G., Rudnitskiy A., A., Soglasnov V., A, Zuga V.
The work describes a system for converting VLBI observation data using the algorithms of coherent dedispersion and compensation of two-bit signal sampling. Coherent dedispersion is important for processing pulsar observations to obtain the best tempo
Externí odkaz:
http://arxiv.org/abs/2303.17280
The dynamical variational autoencoders (DVAEs) are a family of latent-variable deep generative models that extends the VAE to model a sequence of observed data and a corresponding sequence of latent vectors. In almost all the DVAEs of the literature,
Externí odkaz:
http://arxiv.org/abs/2303.09404
Autor:
Yuri N. Girin
Publikováno v:
Литература двух Америк, Iss 16, Pp 103-119 (2024)
The article considers the specific poetry of one of the largest poets of Chile of the 20th century — Pablo de Rokha in relation to the poetry of his compatriot and contemporary Pablo Neruda. Both poets belonged to Latin American avant-garde aesthet
Externí odkaz:
https://doaj.org/article/73dd6cbb245a4604ab8fb4b8b9d47177
Several recent studies have tested the use of transformer language model representations to infer prosodic features for text-to-speech synthesis (TTS). While these studies have explored prosody in general, in this work, we look specifically at the pr
Externí odkaz:
http://arxiv.org/abs/2207.01718
Publikováno v:
Speech Communication, vol. 148, 2023
Understanding and controlling latent representations in deep generative models is a challenging yet important problem for analyzing, transforming and generating various types of data. In speech processing, inspiring from the anatomical mechanisms of
Externí odkaz:
http://arxiv.org/abs/2204.07075