no code implementations • 29 Sep 2021 • Jeremiah Birrell, Markos A. Katsoulakis, Yannis Pantazis, Dipjyoti Paul, Anastasios Tsourtis
Unfortunately, the approximation of expectations that are inherent in variational formulas by statistical averages can be problematic due to high statistical variance, e. g., exponential for the Kullback-Leibler divergence and certain estimators.
1 code implementation • 13 Aug 2020 • Dipjyoti Paul, Muhammed PV Shifas, Yannis Pantazis, Yannis Stylianou
Intelligibility enhancement as quantified by the Intelligibility in Bits (SIIB-Gauss) measure shows that the proposed Lombard-SSDRC TTS system shows significant relative improvement between 110% and 130% in speech-shaped noise (SSN), and 47% to 140% in competing-speaker noise (CSN) against the state-of-the-art TTS approach.
1 code implementation • 9 Aug 2020 • Dipjyoti Paul, Yannis Pantazis, Yannis Stylianou
In terms of performance, our system has been preferred over the baseline TTS system by 60% over 15. 5% and by 60. 9% over 32. 6%, for seen and unseen speakers, respectively.
Ranked #11 on Speech Synthesis on LibriTTS
no code implementations • 11 Jun 2020 • Yannis Pantazis, Dipjyoti Paul, Michail Fasoulakis, Yannis Stylianou, Markos Katsoulakis
In this paper, we propose a novel loss function for training Generative Adversarial Networks (GANs) aiming towards deeper theoretical understanding as well as improved stability and performance for the underlying optimization problem.
no code implementations • 6 Nov 2018 • Yannis Pantazis, Dipjyoti Paul, Michail Fasoulakis, Yannis Stylianou
The impressive success of Generative Adversarial Networks (GANs) is often overshadowed by the difficulties in their training.
no code implementations • 22 Dec 2016 • Monisankha Pal, Dipjyoti Paul, Md Sahidullah, Goutam Saha
Most of the existing studies on voice conversion (VC) are conducted in acoustically matched conditions between source and target signal.