58 citations · 205 across the 130 of their papers we have counts for
21 papers · 1 filter
ED-TTS: Multi-Scale Emotion Modeling using Cross-Domain Emotion Diarization for Emotional Speech Synthesis
Haobin Tang, Xulong Zhang, Ning Cheng +2
Existing emotional speech synthesis methods often utilize an utterance-level style embedding extracted from reference audio, neglecting the inherent multi-scale property of speech…
Boosting Star-GANs for Voice Conversion with Contrastive Discriminator
Shijing Si, Jianzong Wang, Xulong Zhang +3
Nonparallel multi-domain voice conversion methods such as the StarGAN-VCs have been widely applied in many scenarios. However, the training of these models usually poses a challeng…
Speech Representation Disentanglement with Adversarial Mutual Information Learning for One-shot Voice Conversion
SiCheng Yang, Methawee Tantrawenith, Haolin Zhuang +8
One-shot voice conversion (VC) with only a single target speaker's speech for reference has become a hot research topic. Existing works generally disentangle timbre, while informat…
r-G2P: Evaluating and Enhancing Robustness of Grapheme to Phoneme Conversion by Controlled noise introducing and Contextual information incorporation
Chendong Zhao, Jianzong Wang, Xiaoyang Qu +2
Grapheme-to-phoneme (G2P) conversion is the process of converting the written form of words to their pronunciations. It has an important role for text-to-speech (TTS) synthesis and…
Loss Prediction: End-to-End Active Learning Approach For Speech Recognition
Jian Luo, Jianzong Wang, Ning Cheng +1
End-to-end speech recognition systems usually require huge amounts of labeling resource, while annotating the speech data is complicated and expensive. Active learning is the solut…
Dropout Regularization for Self-Supervised Learning of Transformer Encoder Speech Representation
Jian Luo, Jianzong Wang, Ning Cheng +1
Predicting the altered acoustic frames is an effective way of self-supervised learning for speech representation. However, it is challenging to prevent the pretrained model from ov…
Unidirectional Memory-Self-Attention Transducer for Online Speech Recognition
Jian Luo, Jianzong Wang, Ning Cheng +1
Self-attention models have been successfully applied in end-to-end speech recognition systems, which greatly improve the performance of recognition accuracy. However, such attentio…
LVCNet: Efficient Condition-Dependent Modeling Network for Waveform Generation
Zhen Zeng, Jianzong Wang, Ning Cheng +1
In this paper, we propose a novel conditional convolution network, named location-variable convolution, to model the dependencies of the waveform sequence. Different from the use o…
GraphPB: Graphical Representations of Prosody Boundary in Speech Synthesis
Aolan Sun, Jianzong Wang, Ning Cheng +4
This paper introduces a graphical representation approach of prosody boundary (GraphPB) in the task of Chinese speech synthesis, intending to parse the semantic and syntactic relat…
Multi-QuartzNet: Multi-Resolution Convolution for Speech Recognition with Multi-Layer Feature Fusion
Jian Luo, Jianzong Wang, Ning Cheng +2
In this paper, we propose an end-to-end speech recognition network based on Nvidia's previous QuartzNet model. We try to promote the model performance, and design three components:…
End-to-end Silent Speech Recognition with Acoustic Sensing
Jian Luo, Jianzong Wang, Ning Cheng +2
Silent speech interfaces (SSI) has been an exciting area of recent interest. In this paper, we present a non-invasive silent speech interface that uses inaudible acoustic signals t…
A Real-time Robot-based Auxiliary System for Risk Evaluation of COVID-19 Infection
Wenqi Wei, Jianzong Wang, Jiteng Ma +2
In this paper, we propose a real-time robot-based auxiliary system for risk evaluation of COVID-19 infection. It combines real-time speech recognition, temperature measurement, key…
Evolutionary Algorithm Enhanced Neural Architecture Search for Text-Independent Speaker Verification
Xiaoyang Qu, Jianzong Wang, Jing Xiao
State-of-the-art speaker verification models are based on deep learning techniques, which heavily depend on the handdesigned neural architectures from experts or engineers. We borr…
Large-scale Transfer Learning for Low-resource Spoken Language Understanding
Xueli Jia, Jianzong Wang, Zhiyong Zhang +2
End-to-end Spoken Language Understanding (SLU) models are made increasingly large and complex to achieve the state-ofthe-art accuracy. However, the increased complexity of a model…
Prosody Learning Mechanism for Speech Synthesis System Without Text Length Limit
Zhen Zeng, Jianzong Wang, Ning Cheng +1
Recent neural speech synthesis systems have gradually focused on the control of prosody to improve the quality of synthesized speech, but they rarely consider the variability of pr…
MLNET: An Adaptive Multiple Receptive-field Attention Neural Network for Voice Activity Detection
Zhenpeng Zheng, Jianzong Wang, Ning Cheng +2
Voice activity detection (VAD) makes a distinction between speech and non-speech and its performance is of crucial importance for speech based services. Recently, deep neural netwo…
A Comparison of Label-Synchronous and Frame-Synchronous End-to-End Models for Speech Recognition
Linhao Dong, Cheng Yi, Jianzong Wang +4
End-to-end models are gaining wider attention in the field of automatic speech recognition (ASR). One of their advantages is the simplicity of building that directly recognizes the…
MDCNN-SID: Multi-scale Dilated Convolution Network for Singer Identification
Xulong Zhang, Jianzong Wang, Ning Cheng +1
Most singer identification methods are processed in the frequency domain, which potentially leads to information loss during the spectral transformation. In this paper, instead of…
A Robust Speaker Clustering Method Based on Discrete Tied Variational Autoencoder
Chen Feng, Jianzong Wang, Tongxu Li +2
Recently, the speaker clustering model based on aggregation hierarchy cluster (AHC) is a common method to solve two main problems: no preset category number clustering and fix cate…
AlignTTS: Efficient Feed-Forward Text-to-Speech System without Explicit Alignment
Zhen Zeng, Jianzong Wang, Ning Cheng +2
Targeting at both high efficiency and performance, we propose AlignTTS to predict the mel-spectrum in parallel. AlignTTS is based on a Feed-Forward Transformer which generates mel-…
GraphTTS: graph-to-sequence modelling in neural text-to-speech
Aolan Sun, Jianzong Wang, Ning Cheng +3
This paper leverages the graph-to-sequence method in neural text-to-speech (GraphTTS), which maps the graph embedding of the input sequence to spectrograms. The graphical inputs co…