Search

Speech enhancement model is used to map a noisy speech to a clean speech. In the training stage, an objective function is often adopted to optimize the model parameters. However, in most studies, there is an inconsistency between the model optimization criterion and the evaluation criterion on the enhanced speech. For example, in measuring speech intelligibility, most of the evaluation metric is based on a short-time objective intelligibility (STOI) measure, while the frame based minimum mean square error (MMSE) between estimated and clean speech is widely used in optimizing the model. Due to the inconsistency, there is no guarantee that the trained model can provide optimal performance in applications. In this study, we propose an end-to-end utterance-based speech enhancement framework using fully convolutional neural networks (FCN) to reduce the gap between the model optimization and evaluation criterion. Because of the utterance-based optimization, temporal correlation information of long speech segments, or even at the entire utterance level, can be considered when perception-based objective functions are used for the direct optimization. As an example, we implement the proposed FCN enhancement framework to optimize the STOI measure. Experimental results show that the STOI of test speech is better than conventional MMSE-optimized speech due to the consistency between the training and evaluation target. Moreover, by integrating the STOI in model optimization, the intelligibility of human subjects and automatic speech recognition (ASR) system on the enhanced speech is also substantially improved compared to those generated by the MMSE criterion., Comment: Accepted in IEEE Transactions on Audio, Speech and Language Processing (TASLP)

Based on the results of psycholinguistic experiments, the perceptual magnet effect is the important factor in speech development. This effect produced a warped auditory space to the corresponding phoneme. The purpose of this study was to develop a neural network model in simulation of speech perception. The neural network model with unsupervised learning was used to determine the phonetic categories of phoneme according to the formant frequencies of the vowels. The modified Self-Organizing Map (SOM) algorithm was proposed to produce the auditory perceptual space of English vowels. Simulated results were compared with findings from psycholinguistic experiments, such as categorization of English /r/ and /l/ and prototype and non-prototype vowels, to indicate the model's ability to produce auditory perception space. In addition, this speech perception model was combined with the neural network model (Directions into Velocities Articulator, DIVA) to simulate categorization of ten English vowels and their prod...