Zobrazeno 1 - 10
of 1 128
pro vyhledávání: '"Homayoon, P."'
Autor:
Liu, Xing Yi, Beigi, Homayoon
Publikováno v:
Recognition Technologies, Inc. Technical Report, 2024
Presently, punctuation restoration models are evaluated almost solely on well-structured, scripted corpora. On the other hand, real-world ASR systems and post-processing pipelines typically apply towards spontaneous speech with significant irregulari
Externí odkaz:
http://arxiv.org/abs/2409.11241
Autor:
Ter-Avanesov, Boris, Beigi, Homayoon
We explore the performance of various artificial neural network architectures, including a multilayer perceptron (MLP), Kolmogorov-Arnold network (KAN), LSTM-GRU hybrid recursive neural network (RNN) models, and a time-delay neural network (TDNN) for
Externí odkaz:
http://arxiv.org/abs/2409.06724
Second-order information is valuable for many applications but challenging to compute. Several works focus on computing or approximating Hessian diagonals, but even this simplification introduces significant additional costs compared to computing a g
Externí odkaz:
http://arxiv.org/abs/2406.03276
Autor:
Natesan, Sanjay, Beigi, Homayoon
Publikováno v:
Recognition Technologies, Inc. Technical Report (2024), RTI-20240524-01
Large scale machine learning-based Raga identification continues to be a nontrivial issue in the computational aspects behind Carnatic music. Each raga consists of many unique and intrinsic melodic patterns that can be used to easily identify them fr
Externí odkaz:
http://arxiv.org/abs/2405.16000
Market and user characteristics of mobile apps make their release management different from proprietary software products and web services. Despite the wealth of information regarding users' feedback on an app, an in-depth analysis of app releases is
Externí odkaz:
http://arxiv.org/abs/2405.15752
The increasing attraction of mobile apps has inspired researchers to analyze apps from different perspectives. As with any software product, apps have different attributes such as size, content maturity, rating, category, or number of downloads. Curr
Externí odkaz:
http://arxiv.org/abs/2405.15737
Polar codes are the first class of structured channel codes that achieve the symmetric capacity of binary channels with efficient encoding and decoding. In 2019, Arikan proposed a new polar coding scheme referred to as polarization-adjusted convoluti
Externí odkaz:
http://arxiv.org/abs/2401.13922
Publikováno v:
Recognition Technologies, Inc. Technical Report (2023), RTI-20230328-01
Most state-of-the-art spoken language identification models are closed-set; in other words, they can only output a language label from the set of classes they were trained on. Open-set spoken language identification systems, however, gain the ability
Externí odkaz:
http://arxiv.org/abs/2308.14951
Autor:
Salehi, Mahdi, Oraizi, Homayoon
This study presents the design of four $178\times178$ $(mm)^{2}$ wideband, high gain, highly efficient metasurface-based 4T4R MIMO (Multiple-Input Multiple-Output) antennas with highly isolated ports, covering the middle and a portion of the upper ba
Externí odkaz:
http://arxiv.org/abs/2309.07131
Autor:
Farrahi, Homayoon, Mahmood, A. Rupam
Continuous-time reinforcement learning tasks commonly use discrete steps of fixed cycle times for actions. As practitioners need to choose the action-cycle time for a given task, a significant concern is whether the hyper-parameters of the learning a
Externí odkaz:
http://arxiv.org/abs/2305.05760