Prosodic Representations of Prominence Classification Neural Networks and Autoencoders Using Bottleneck Features

Abstract

Prominence perception has been known to correlate with a complex interplay of the acoustic features of energy, fundamental frequency, spectral tilt, and duration. The contribution and importance of each of these features in distinguishing between prominent and non-prominent units in speech is not always easy to determine, and more so, the prosodic representations that humans and automatic classifiers learn have been difficult to interpret. This work focuses on examining the acoustic prosodic representations that binary prominence classification neural networks and autoencoders learn for prominence. We investigate the complex features learned at different layers of the network as well as the 10-dimensional bottleneck features (BNFs), for the standard acoustic prosodic correlates of prominence separately and in combination. We analyze and visualize the BNFs obtained from the prominence classification neural networks as well as their network activations. The experiments are conducted on a corpus of Dutch continuous speech with manually annotated prominence labels. Our results show that the prosodic representations obtained from the BNFs and higher-dimensional non-BNFs provide good separation of the two prominence categories, with, however, different partitioning of the BNF space for the distinct features, and the best overall separation obtained for F0.

On Training Targets and Activation Functions for Deep Representation Learning in Text-Dependent Speaker Verification

“…Activation functions, on the other hand, control the output of DNN hidden neurons, as well as the gradient contribution during the error back-propagation process for network parameter optimization. Among others, sigmoid [24,25] and ReLU [26] are most widely used in the state-of-the-art systems such as speaker recognition [27][28][29][30] and language recognition [31,32], speech recognition [33,34], prosodic representation [35], and image processing [19,22,23].…”

On Training Targets and Activation Functions for Deep Representation Learning in Text-Dependent Speaker Verification

“…Activation functions, on the other hand, control the output of DNN hidden neurons as well as the gradient contribution during the error back-propagation process for network parameter optimization. Among others, sigmoid [19], [20] and ReLU [21] are most widely used in the state-of-the-art systems such as speaker recognition [22], [23], [24], [25] and language recognition [26], [27], speech recognition [28], [29], prosodic representation [30] and image processing [14], [17], [18].…”