Zobrazeno 1 - 10
of 243
pro vyhledávání: '"Trần, Dũng"'
Existing zero-shot text-to-speech (TTS) systems are typically designed to process complete sentences and are constrained by the maximum duration for which they have been trained. However, in many streaming applications, texts arrive continuously in s
Externí odkaz:
http://arxiv.org/abs/2410.00767
LiveSpeech: Low-Latency Zero-shot Text-to-Speech via Autoregressive Modeling of Audio Discrete Codes
Prior works have demonstrated zero-shot text-to-speech by using a generative language model on audio tokens obtained via a neural audio codec. It is still challenging, however, to adapt them to low-latency scenarios. In this paper, we present LiveSpe
Externí odkaz:
http://arxiv.org/abs/2406.02897
Masked Autoencoders (MAEs) learn rich low-level representations from unlabeled data but require substantial labeled data to effectively adapt to downstream tasks. Conversely, Instance Discrimination (ID) emphasizes high-level semantics, offering a po
Externí odkaz:
http://arxiv.org/abs/2403.09579
Learned image compression has gained widespread popularity for their efficiency in achieving ultra-low bit-rates. Yet, images containing substantial textual content, particularly screen-content images (SCI), often suffers from text distortion at such
Externí odkaz:
http://arxiv.org/abs/2402.08643
Generalization remains a major problem in supervised learning of single-channel speech enhancement. In this work, we propose learnable loss mixup (LLM), a simple and effortless training diagram, to improve the generalization of deep learning-based sp
Externí odkaz:
http://arxiv.org/abs/2312.17255
Autor:
Lin, Chun-Chi, Tran, Dung The
In this paper, we resolve problems of spline interpolation, regardless of whether least-squares fitting is incorporated, on smooth Riemannian manifolds. Our approach leverages the concept of gradient flows for successively connected curves or network
Externí odkaz:
http://arxiv.org/abs/2312.10513
Diffusion models are instrumental in text-to-audio (TTA) generation. Unfortunately, they suffer from slow inference due to an excessive number of queries to the underlying denoising network per generation. To address this bottleneck, we introduce Con
Externí odkaz:
http://arxiv.org/abs/2309.10740
Autor:
Remke, Anne, Tran, Dung Hoang
Publikováno v:
EPTCS 361, 2022
The proceedings of the 7th International Workshop on Symbolic-Numeric Methods for Reasoning about CPS and IoT (SNR 2021) feature five peer-reviewed contributions and three invited talks. SNR focuses on the combination of symbolic and numeric methods
Externí odkaz:
http://arxiv.org/abs/2207.04391
The recent breakthrough of Guth, Iosevich, Ou, and Wang (2019) on the Falconer distance problem states that for a compact set $A\subset \mathbb{R}^2$, if the Hausdorff dimension of $A$ is greater than $\frac{5}{4}$, then the distance set $\Delta(A)$
Externí odkaz:
http://arxiv.org/abs/2203.10423
Autor:
Tran, Dung Duc, Park, Edward, Thu Van, Can, Nguyen, Thien Duc, Nguyen, Au Hai, Linh, Tran Che, Quyen, Pham Hong, Tran, Duong Anh, Nguyen, Hong Quan
Publikováno v:
In Heliyon 15 September 2024 10(17)