Chang ZENG
Chang ZENG
Home
News
Publications
Skills
Activities
Experience
Posts
Light
Dark
Automatic
3
Uni-Stream: A Unified Streaming Speech Language Model for Speech Synthesis and Recognition
Uni-Stream unifies real-time speech synthesis and recognition in one streaming speech language model and analyzes its task- and modality-specific representations.
Xuan Shi
,
Chang Zeng
,
Tiantian Feng
,
Ye Zhang
,
Jun Liu
,
Shih-Heng Wang
,
Lie Lu
,
Shrikanth Narayanan
PDF
Cite
DOI
ArXiv
Fish Audio Drama 3 Technical Report
Fish Audio Drama 3 is an open-source speech generation model for directed multi-character performances, iterative revision, speech infilling, and long-form or streaming synthesis.
Dawei Han
,
Shidong Li
,
Shijia Liao
,
Jiahua Liu
,
Songting Liu
,
Qingzheng Wang
,
Yuxuan Wang
,
Xin Chen
,
Yifan Cheng
,
Tianyu Li
,
Gangin Park
,
Xuehao Yang
,
Chang Zeng
,
Haoyang Zhang
,
Ruoyi Zhang
,
Xiangyu Zhang
PDF
Cite
Code
Model
ArXiv
Can LLM Agents Automate Reinforcement Learning for Text-to-Speech?
AgenticTTS-Forge studies automated reinforcement-learning post-training for TTS and identifies measurement, reward-design, and policy-composition challenges for LLM research agents.
Xuanjun Chen
,
Zixiong Su
,
Hao Shi
,
Chang Zeng
,
Kai Li
,
Jyh-Shing Roger Jang
,
Hung-yi Lee
PDF
Cite
ArXiv
UNITE-AUDIO: Joint Learning of Continuous Tokenization and Latent Flow Matching for Text-to-Audio Generation
UNITE-AUDIO jointly learns continuous audio tokenization and latent flow matching, with Flow-GRPO post-training for compact text-to-audio generation.
Runwu Shi
,
Kai Li
,
Yujin Wang
,
Dong Yang
,
Jiahui Li
,
Jiang Wang
,
Chang Zeng
,
Benjamin Yen
,
Ashizawa Takeshi
,
Chunxiang Jin
,
Kazuhiro Nakadai
PDF
Cite
Project
ArXiv
StepAudio 2.5 Technical Report
StepAudio 2.5 is a unified audio-language foundation model for ASR, TTS, and realtime spoken interaction with task-tailored RLHF and specialized decoding.
Bin Lin
,
Bo Zhao
,
Boyong Wu
,
Chao Yan
,
Chen Wu
,
Chang Zeng
,
et al.
PDF
Cite
ArXiv
BEAVER: A Training-Free Hierarchical Prompt Compression Method via Structure-Aware Page Selection
BEAVER is a training-free hierarchical prompt compression framework for long-context LLM inference with strong benchmark performance and 26.4x lower latency on 128k contexts.
Zhengpei Hu
,
Kai Li
,
Dapeng Fu
,
Chang Zeng
,
Yue Li
,
Yuanhao Tang
,
Jianqiang Huang
PDF
Code
Project
Demo
ArXiv
Towards Interactive Intelligence for Digital Humans
Mio is an end-to-end multimodal interactive digital-human framework that combines reasoning and real-time embodiment with state-of-the-art performance.
Yiyi Cai
,
Xuangeng Chu
,
Xiwei Gao
,
Sitong Gong
,
Yifei Huang
,
Caixin Kang
,
Kunhang Li
,
Haiyang Liu
,
Ruicong Liu
,
Yun Liu
,
Dianwen Ng
,
Zixiong Su
,
Erwin Wu
,
Yuhan Wu
,
Dingkun Yan
,
Tianyu Yan
,
Chang Zeng
,
Bo Zheng
,
You Zhou
PDF
Project
ArXiv
Demo
HAM-TTS: Hierarchical Acoustic Modeling for Token-Based Zero-Shot Text-to-Speech with Model and Data Scaling
We introduce a novel token-based text-to-speech (TTS) model with 0.8B parameters, trained on a mix of real and synthetic data totaling 650k hours, to address issues like pronunciation accuracy and style consistency. This model integrates a latent variable sequence with enhanced acoustic information into the TTS system, reducing errors and style changes. Our training includes data augmentation for improved timbre consistency, and we use a few-shot voice conversion model to generate diverse voices. This approach enables learning of one-to-many mappings in speech, ensuring both diversity and timbre consistency. Our model outperforms VALL-E in pronunciation, style maintenance, and timbre continuity.
Chunhui Wang
,
Chang Zeng
,
Bowen Zhang
,
Ziyang Ma
,
Yefan Zhu
,
Zifeng Cai
,
Jian Zhao
,
Zhonglin Jiang
,
Yong Chen
PDF
Cite
Project
DOI
ArXiv
Cite
×