arXiv subjects

Sheng Zhao

Publications and source records attributed to Sheng Zhao.

93 records · Page 6Linked to original sources

arXiv2019

Co-optimization of a piezoelectric energy harvesting system for broadband operation

The goal of this research is to increase the bandwidth (BW) over which substantial energy can be harvested using a piezoelectric energy harvester (PEH). The key innovation is the use of bias-flip (BF) electronics at the output of a PEH having a large electromechanical coupling coefficient $κ_{e}^{2}$. For a PEH with large $κ_{e}^{2}$, the open-circuit resonance frequency $f_{oc}$ is substantially larger than the short-circuit resonance frequency $f_{sc}$. Over the intervening range, the reactive part of the conjugate matched load impedance is small, and can be approximated using BF electronics in which the BF voltage is sufficiently small and the BF losses are small. This results in a large BW over which substantial energy can be harvested. Experimental results using a commercially available PEH are presented to demonstrate this concept. Design guidelines are provided for achieving PEHs having increased $κ_{e}^{2}$.

eess.SP↗

arXiv2019

Token-Level Ensemble Distillation for Grapheme-to-Phoneme Conversion

Grapheme-to-phoneme (G2P) conversion is an important task in automatic speech recognition and text-to-speech systems. Recently, G2P conversion is viewed as a sequence to sequence task and modeled by RNN or CNN based encoder-decoder framework. However, previous works do not consider the practical issues when deploying G2P model in the production system, such as how to leverage additional unlabeled data to boost the accuracy, as well as reduce model size for online deployment. In this work, we propose token-level ensemble distillation for G2P conversion, which can (1) boost the accuracy by distilling the knowledge from additional unlabeled data, and (2) reduce the model size but maintain the high accuracy, both of which are very practical and helpful in the online production system. We use token-level knowledge distillation, which results in better accuracy than the sequence-level counterpart. What is more, we adopt the Transformer instead of RNN or CNN based models to further boost the accuracy of G2P conversion. Experiments on the publicly available CMUDict dataset and an internal English dataset demonstrate the effectiveness of our proposed method. Particularly, our method achieves 19.88% WER on CMUDict dataset, outperforming the previous works by more than 4.22% WER, and setting the new state-of-the-art results.

cs.CL↗

arXiv2019

Neural Speech Synthesis with Transformer Network

Although end-to-end neural text-to-speech (TTS) methods (such as Tacotron2) are proposed and achieve state-of-the-art performance, they still suffer from two problems: 1) low efficiency during training and inference; 2) hard to model long dependency using current recurrent neural networks (RNNs). Inspired by the success of Transformer network in neural machine translation (NMT), in this paper, we introduce and adapt the multi-head attention mechanism to replace the RNN structures and also the original attention mechanism in Tacotron2. With the help of multi-head self-attention, the hidden states in the encoder and decoder are constructed in parallel, which improves the training efficiency. Meanwhile, any two inputs at different times are connected directly by self-attention mechanism, which solves the long range dependency problem effectively. Using phoneme sequences as input, our Transformer TTS network generates mel spectrograms, followed by a WaveNet vocoder to output the final audio results. Experiments are conducted to test the efficiency and performance of our new network. For the efficiency, our Transformer TTS network can speed up the training about 4.25 times faster compared with Tacotron2. For the performance, rigorous human tests show that our proposed model achieves state-of-the-art performance (outperforms Tacotron2 with a gap of 0.048) and is very close to human quality (4.39 vs 4.44 in MOS).

cs.CL↗