Exponential moving average model in parallel speech recognition training

[PDF] arxiv.org

Deep LSTM for large vocabulary continuous speech recognition

X Tian, J Zhang, Z Ma, Y He, J Wei, P Wu… - arXiv preprint arXiv …, 2017 - arxiv.org
Recurrent neural networks (RNNs), especially long short-term memory (LSTM) RNNs, are
effective network for sequential task like speech recognition. Deeper LSTM models perform …
被引用次数：27 相关文章 所有 2 个版本
[PDF] arxiv.org

Frame stacking and retaining for recurrent neural network acoustic model

X Tian, J Zhang, Z Ma, Y He, J Wei - arXiv preprint arXiv:1705.05992, 2017 - arxiv.org
Frame stacking is broadly applied in end-to-end neural network training like connectionist
temporal classification (CTC), and it leads to more accurate models and faster decoding …
被引用次数：6 相关文章 所有 2 个版本
[PDF] unad.edu.co

Diseño y fabricación de regulador MPPT de bajo costo

HE Segura Garcia, R Valencia Pastrana - repository.unad.edu.co
La energía solar fotovoltaica es clasificada como una fuente de energía renovable,
caracterizado por un mantenimiento muy reducido y en muchas ocasiones a bajo costo …
相关文章

High Dimensional Time Series Classification Based on Multi-Layer Perceptron and Moving Average Model

J Li, C Xu, T Zhang - 2019 Chinese Automation Congress …, 2019 - ieeexplore.ieee.org
In order to improve the performance of high dimensional time series classification, a High
Dimensional Time Series Classification Model (HDTSCM) based on multi-layer perceptron …
相关文章