This paper proposes a lattice-based sequential discriminative training method to extract more discriminative bottleneck features.

ABSTRACT. This paper proposes a lattice-based sequential discriminative train- ing method to extract more discriminative bottleneck features. In.

In this method, the bottleneck neural network is first trained with cross entropy criteria, and then only the weights of bottleneck layer are retrained with ...

This paper proposes a lattice-based sequential discriminative training method to extract more discriminative bottleneck features. In our method, the bottleneck ...

Lattice based optimization of bottleneck feature extractor with linear transformation. D. Liu, S. Wei, W. Guo, Y. Bao, S. Xiong, and L. Dai.

This paper investigates a method for training bottleneck. (BN) features in a more targeted manner for their intended use in GMM-HMM based ASR.

This paper investigates a method for training bottleneck (BN) features in a more targeted manner for their intended use in GMM-HMM based ASR, and shows that ...

This paper investigates a method for training bottleneck (BN) features in a more targeted manner for their intended use in GMM-HMM based ASR.

Diyuan Liu, Si Wei, Wu Guo, Yebo Bao, Shifu Xiong, Li-Rong Dai: Lattice based optimization of bottleneck feature extractor with linear transformation.