A neural network model that significant improves unitselection-based Text-To-Speech synthesis is presented. The model employs a sequence-to-sequence LSTM-based autoencoder that compresses the acoustic and linguistic features of each unit to a fixed-size vector referred to as an embedding. Unit-selection is facilitated by formulating the target cost as an L2 distance in the embedding space. In o...

In this paper, a pitch-adaptive waveform generative model named Quasi-Periodic WaveNet (QPNet) is proposed to improve the limited pitch controllability of vanilla (WN) using pitch-dependent dilated convolution neural networks (PDCNNs). Specifically, as probabilistic autoregressive generation with stacked layers, WN achieves high-fidelity audio generation. However, pure-data-driven nature and la...

Due to the increasing need to distributed energy resources in power systems, their problems should be studied. One of the main problem of distributed energy resources is unplanned islanding. The unplanned islanding has some dangers to the power systems and the repairman which are works with the incorrect devices. In this paper, a passive local method is proposed. The proposed method is based on...

Text-to-Speech (TTS) system converts an input text into a synthetic speech output. The paper provides detailed description of developing Kiswahili TTS using Tacotron 2 architecture and WaveNet vocoder. will help visually impaired persons to learn, assist with communication disorders, provide source in Voice Alarm Public Address equipments. is sequence-to-sequence model for building systems. con...

In this paper, we propose a new framework for environmental sound synthesis using onomatopoeic words and event labels. The conventional method of synthesis, in which only labels are used, cannot finely control the time-frequency structural features synthesized sounds, such as duration, timbre, pitch. There various ways to express other than labels, use words. An word, is character sequence phon...

Recorded speech signals often contain noise that affects the quality of signal and reduces intelligibility. Several studies have used Generative Adversarial Networks (GANs) to remove artifacts improve However, GANs can suffer from gradient vanishing or explosion reduce their effectiveness in denoising. To mitigate vanishing, we applied CoulombGAN architecture denoising using a model structure s...

We present a new model for singing synthesis based on a modified version of the WaveNet architecture. Instead of modeling raw waveform, we model features produced by a parametric vocoder that separates the influence of pitch and timbre. This allows conveniently modifying pitch to match any target melody, facilitates training on more modest dataset sizes, and significantly reduces training and g...

Most existing neural network models for music generation use recurrent neural networks. However, the recent WaveNet model proposed by DeepMind shows that convolutional neural networks (CNNs) can also generate realistic musical waveforms in the audio domain. Following this light, we investigate using CNNs for generating melody (a series of MIDI notes) one bar after another in the symbolic domain...