Extended VTS for noise-robust speech recognition

43 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Nonlinear Compensation Using the Gauss–Newton Method for Noise-Robust Speech Recognition (2012, background): “It is not straightforward to extend the sampling-based methods to the dynamic parameters in a principled way, though possible [32], [33].”
Improving Robustness of Deep Neural Network Acoustic Models via Speech Separation and Joint Adaptive Training (2015, methodology): “For example, delta features can be calculated using a linear activation layer with weights as below [39]: Here, is the static feature at time , is the identity matrix, and is the weight matrix of the hidden layer.”
An Overview of Noise-Robust Automatic Speech Recognition (2014, methodology): “(107) is used in uncertainty decoding with SPLICE [239] and joint uncertainty decoding work [192], [212], [240], [241] which we will discuss in detail in the next section.”
Joint Uncertainty Decoding for Noise Robust Subspace Gaussian Mixture Models (2013, methodology): “Experiments on the Aurora 4 corpus are reported and analysed in Section VI, followed by a discussion and conclusions in Section VII.”
Extended VTS for Noise-Robust Speech Recognition (2011, methodology): “This is discussed in more detail in [13].”
Joint noise adaptive training for robust automatic speech recognition (2014, methodology): “For example, delta features can be calculated using a linear activation layer with weights as below [24]: [ ot △ot ]”