Zobrazeno 1 - 10
of 561
pro vyhledávání: '"WEISS, RON"'
Autor:
Kermanshah, Mehdi, Beaver, Logan E., Sokolich, Max, Kirmizitas, Fatma Ceren, Das, Sambeeta, Tron, Roberto, Weiss, Ron, Belta, Calin
This paper presents a control framework for magnetically actuated cellbots, which combines Model Predictive Control (MPC) with Gaussian Processes (GPs) as a disturbance estimator for precise trajectory tracking. To address the challenges posed by unm
Externí odkaz:
http://arxiv.org/abs/2406.02722
Micron-scale robots ($\mu$bots) have recently shown great promise for emerging medical applications. Accurate controlling $\mu$bots, while critical to their successful deployment, is challenging. In this work, we consider the problem of tracking a re
Externí odkaz:
http://arxiv.org/abs/2212.00188
Autor:
Wang, Gary, Cubuk, Ekin D., Rosenberg, Andrew, Cheng, Shuyang, Weiss, Ron J., Ramabhadran, Bhuvana, Moreno, Pedro J., Le, Quoc V., Park, Daniel S.
Data augmentation is a ubiquitous technique used to provide robustness to automatic speech recognition (ASR) training. However, even as so much of the ASR training process has become automated and more "end-to-end", the data augmentation policy (what
Externí odkaz:
http://arxiv.org/abs/2210.10879
Publikováno v:
PMLR 168:968-980, 2022
Learning dynamical systems properties from data provides important insights that help us understand such systems and mitigate undesired outcomes. In this work, we propose a framework for learning spatio-temporal (ST) properties as formal logic specif
Externí odkaz:
http://arxiv.org/abs/2112.10714
Autor:
Chen, Nanxin, Zhang, Yu, Zen, Heiga, Weiss, Ron J., Norouzi, Mohammad, Dehak, Najim, Chan, William
This paper introduces WaveGrad 2, a non-autoregressive generative model for text-to-speech synthesis. WaveGrad 2 is trained to estimate the gradient of the log conditional density of the waveform given a phoneme sequence. The model takes an input pho
Externí odkaz:
http://arxiv.org/abs/2106.09660
Supervised neural network training has led to significant progress on single-channel sound separation. This approach relies on ground truth isolated sources, which precludes scaling to widely available mixture data and limits progress on open-domain
Externí odkaz:
http://arxiv.org/abs/2106.00847
Autor:
de Almeida Magalhaes, Taciani, Liu, Jingjing, Chan, Charlene, Borges, Kleiton Silva, Zhang, Jiuchun, Kane, Andrew J., Wierbowski, Bradley M., Ge, Yunhui, Liu, Zhiwen, Mannam, Prabhath, Zeve, Daniel, Weiss, Ron, Breault, David T., Huang, Pengxiang, Salic, Adrian
Publikováno v:
In Developmental Cell 22 January 2024 59(2):244-261
We describe a sequence-to-sequence neural network which directly generates speech waveforms from text inputs. The architecture extends the Tacotron model by incorporating a normalizing flow into the autoregressive decoder loop. Output waveforms are m
Externí odkaz:
http://arxiv.org/abs/2011.03568
We propose a multitask training method for attention-based end-to-end speech recognition models. We regularize the decoder in a listen, attend, and spell model by multitask training it on both audio-text and text-only data. Trained on the 100-hour su
Externí odkaz:
http://arxiv.org/abs/2010.14318
Although neural end-to-end text-to-speech models can synthesize highly natural speech, there is still room for improvements to its efficiency and naturalness. This paper proposes a non-autoregressive neural text-to-speech model augmented with a varia
Externí odkaz:
http://arxiv.org/abs/2010.11439