吴志勇｜湖北大学｜学者主页

2026

FineCombo-TTS: Collaborative and Precise Controllable Speech Synthesis Using Text Descriptions and Reference Speech

Shuo-Yi Zhou, Yixuan Zhou, Pei-Yin Yang, Yifan Hu, Yicheng Zhong, Zhi-Sheng Wang, Zhiyong Wu

arXiv.org

2026

SonicWeave: Chunk-Routed Mixture-of-Experts for Unified Audio Scene Generation

Yun-Rui Cai, Xu Li, Yucheng Zhou, Jin-Chao Li, Ding-Dong Wang, Dongchao Yang, Xi-Xin Wu, Chen Zhang, Zhiyong Wu, Peng-Fei Wan, Helen M. Meng

2026

Qwen-Music Technical Report

Jin Xu, Kangdi Wang, Rui-Bin Yuan, Shun-Wei Lei, Xiong Wang, Xi-Ze Cheng, Xueyao Zhang, Yang Zhang, Yiheng Chen, Yongqi Wang, Yue Wang, Zhi-Fang Guo, Zi-Han Liu, Zi-Jian Lin, Da-Ke Guo, Hang-Rui Hu, Lei Xie, Lin-Han Ma, Wei Xue, Wenxiang Guo, Xin-Fa Zhu, Xipin Wei, Yang-Ze Li, Yuanjun Lv, Yuxuan Wang, Yun-Fei Chu, Zhi-Yong Wu

arXiv.org

精选作品

期刊论文

FineCombo-TTS: Collaborative and Precise Controllable Speech Synthesis Using Text Descriptions and Reference Speech

SonicWeave: Chunk-Routed Mixture-of-Experts for Unified Audio Scene Generation

Qwen-Music Technical Report

BridgeCodec: Mamba Enhanced Neural Audio Codec with Schr\"odinger Bridge at Low Bitrate

SongBench: A Fine-Grained Multi-Aspect Benchmark for Song Quality Assessment

Audio Tokens as a Budgeted Resource: Marginal-Utility Allocation for Scalable Audio Representations

UniStream: Multi-Expert Residual Vector Quantization for 48 kHz Causal Streaming Audio Coding

BAMU: Bitstream-Aware Marginal-Utility Allocation for Frozen Pretrained Neural Speech Codecs

SPG-Codec: Exploring the Role and Boundaries of Semantic Priors in Ultra-Low-Bitrate Neural Speech Coding

HarmoniVox: Painting Voices to Match the Avatar's Soul

A Dual-Branch Ensemble Framework for Personality Recognition Based on Multimodal Emotion Features

DualSpeechLM: Towards Unified Speech Understanding and Generation via Dual Speech Token Modeling with Large Language Models

Inferring Speaking Styles for Conversational Speech Synthesis by Learning Contextual Dependencies

Echo: Enhancing Conversational Behavior Generation via Hierarchical Semantic Comprehension with Large Language Models

MuCodec: Ultra Low-Bitrate Music Codec for Music Generation

DAFMSVC: One-Shot Singing Voice Conversion with Dual Attention Mechanism and Flow Matching

DiffCSS: Diverse and Expressive Conversational Speech Synthesis with Diffusion Models

Singing Voice Conversion with Accompaniment Using Self-Supervised Representation-Based Melody Features

NLPP: A Natural Language Prosodic Prominence Dataset Assisted by ChatGPT

LoRA-MER: Low-Rank Adaptation of Pre-Trained Speech Models for Multimodal Emotion Recognition Using Mutual Information

ERVQ: Leverage Residual Vector Quantization for Speech Emotion Recognition

VoxInstruct: Expressive Human Instruction-to-Speech Generation with Unified Multilingual Codec Language Modelling

Inferring Agent Speaking Styles for Auditory-Visual User-Agent Conversation

CMAST: Efficient Speech-Text Joint Training Method to Enhance Linguistic Features Learning of Speech Representations

SpeechCraft: A Fine-Grained Expressive Speech Dataset with Natural Language Description

The Codec Language Model-Based Zero-Shot Spontaneous Style TTS System for CoVoC Challenge 2024

RobustSVC: HuBERT-Based Melody Extractor and Adversarial Learning for Robust Singing Voice Conversion

What Does Your Face Sound Like? 3D Face Shape towards Voice

Hiformer: Sequence Modeling Networks With Hierarchical Attention Mechanisms

DiffuseStyleGesture: Stylized Audio-Driven Co-Speech Gesture Generation with Diffusion Models

Prosody Modeling with 3D Visual Information for Expressive Video Dubbing

VAST: Vivify Your Talking Avatar via Zero-Shot Expressive Facial Style Transfer

A Discourse-level Multi-scale Prosodic Model for Fine-grained Emotion Analysis

QPGesture: Quantization-Based and Phase-Guided Motion Matching for Natural Speech-Driven Gesture Generation

Speaker Characteristics Guided Speech Synthesis

A Compact Framework for Voice Conversion Using Wavenet Conditioned on Phonetic Posteriorgrams

Speech Emotion Recognition Using Capsule Networks

End-to-end Code-switched TTS with Mix of Monolingual Recordings

NN-based Ordinal Regression for Assessing Fluency of ESL Speech

Learning Discriminative Features from Spectrograms Using Center Loss for Speech Emotion Recognition

Dilated Residual Network with Multi-head Self-attention for Speech Emotion Recognition

Quasi-fully Convolutional Neural Network with Variational Inference for Speech Synthesis

Unsupervised Discovery of an Extended Phoneme Set in L2 English Speech for Mispronunciation Detection and Diagnosis

Integrating Articulatory Features into Acoustic-Phonemic Model for Mispronunciation Detection and Diagnosis in L2 English Speech

Learning Frame-Level Recurrent Neural Networks Representations for Query-by-Example Spoken Term Detection on Mobile Devices

Detection of Glottal Closure Instants from Speech Signals: A Convolutional Neural Network Based Method

Feature Based Adaptation for Speaking Style Synthesis

Emphatic Speech Generation with Conditioned Input Layer and Bidirectional LSTMS for Expressive Speech Synthesis

Inferring User Emotive State Changes in Realistic Human-Computer Conversational Dialogs

Emphatic Speech Synthesis and Control Based on Characteristic Transferring in End-to-End Speech Synthesis

Siamese Recurrent Auto-Encoder Representation for Query-by-Example Spoken Term Detection

Emotion Recognition from Variable-Length Speech Segments Using Deep Learning on Spectrograms

Applying Multitask Learning to Acoustic-Phonemic Model for Mispronunciation Detection and Diagnosis in L2 English Speech

Multi-modal Multi-scale Speech Expression Evaluation in Computer-Assisted Language Learning

Automatic lexical stress and pitch accent detection for L2 English speech using multi-distribution deep neural networks

Rapid Style Adaptation Using Residual Error Embedding for Expressive Speech Synthesis

Speech Super-Resolution Using Parallel WaveNet

Attention-based Recurrent Generator with Gaussian Tolerance for Statistical Parametric Speech Synthesis

Multi-Task Learning for Prosodic Structure Generation Using BLSTM RNN with Structured Output Layer

Learning cross-lingual knowledge with multilingual BLSTM for emphasis detection with limited training data

Spectro-Temporal Modelling with Time-Frequency LSTM and Structured Output Layer for Voice Conversion

Multi-task learning of structured output layer bidirectional LSTMS for speech synthesis

Speech Emotion Recognition with Emotion-Pair Based Framework Considering Emotion Distribution Information in Dimensional Emotion Space

Multi-Task Deep Learning for User Intention Understanding in Speech Interaction Systems

Recognizing stances in Mandarin social ideological debates with text and acoustic features

DBLSTM-based multi-task learning for pitch transformation in voice conversion

Question detection from acoustic features using recurrent neural network with gated recurrent unit

Learning cross-lingual information with multilingual BLSTM for speech synthesis of low-resource languages

Analysis on Gated Recurrent Unit Based Question Detection Approach

Combining CNN and BLSTM to Extract Textual and Acoustic Features for Recognizing Stances in Mandarin Ideological Debate Competition

Expressive Speech Driven Talking Avatar Synthesis with DBLSTM Using Limited Amount of Emotional Bimodal Data

Low level descriptors based DBLSTM bottleneck feature for speech driven talking avatar

Study on Feature Subspace of Archetypal Emotions for Speech Emotion Recognition

Phoneme Embedding and its Application to Speech Driven Talking Avatar Synthesis

Using tilt for automatic emphasis detection with Bayesian networks

Understanding speaking styles of internet speech data with LSTM and low-resource training

Acoustic to articulatory mapping with deep neural network

A deep recurrent approach for acoustic-to-articulatory inversion