Cover Image

On the Distillation Loss Functions of Speech VAE for Unified Reconstruction, Understanding, and Generation

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Cheng, Changhao, Wang, Wei, Zhang, Wangyou, Jia, Dongya, Wu, Jian, Chen, Zhuo, Qian, Yanmin
Format:	Preprint
Published:	2026
Subjects:	Sound
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Advanced Zero-Shot Text-to-Speech for Background Removal and Preservation with Controllable Masked Speech Prediction
by: Zhang, Leying, et al.
Published: (2025)

Scale This, Not That: Investigating Key Dataset Attributes for Efficient Speech Enhancement Scaling
by: Zhang, Leying, et al.
Published: (2024)

Improving Design of Input Condition Invariant Speech Enhancement
by: Zhang, Wangyou, et al.
Published: (2024)

Vevo2: A Unified and Controllable Framework for Speech and Singing Voice Generation
by: Zhang, Xueyao, et al.
Published: (2025)

Improving Speech Enhancement with Multi-Metric Supervision from Learned Quality Assessment
by: Wang, Wei, et al.
Published: (2025)

Toward Universal Speech Enhancement for Diverse Input Conditions
by: Zhang, Wangyou, et al.
Published: (2023)

Beyond Performance Plateaus: A Comprehensive Study on Scalability in Speech Enhancement
by: Zhang, Wangyou, et al.
Published: (2024)

Representation-Regularized Convolutional Audio Transformer for Audio Understanding
by: Han, Bing, et al.
Published: (2026)

UrgentMOS: Unified Multi-Metric and Preference Learning for Robust Speech Quality Assessment
by: Wang, Wei, et al.
Published: (2026)

Training Text-to-Speech Model with Purely Synthetic Data: Feasibility, Sensitivity, and Generalization Capability
by: Zhou, Tingxiao, et al.
Published: (2025)

DiTAR: Diffusion Transformer Autoregressive Modeling for Speech Generation
by: Jia, Dongya, et al.
Published: (2025)

DQ-Whisper: Joint Distillation and Quantization for Efficient Multilingual Speech Recognition
by: Shao, Hang, et al.
Published: (2023)

ICASSP 2026 URGENT Speech Enhancement Challenge
by: Li, Chenda, et al.
Published: (2026)

SLM-SS: Speech Language Model for Generative Speech Separation
by: Li, Tianhua, et al.
Published: (2026)

Cross-Utterance Conditioned VAE for Speech Generation
by: Li, Yang, et al.
Published: (2023)

SpeechJudge: Towards Human-Level Judgment for Speech Naturalness
by: Zhang, Xueyao, et al.
Published: (2025)

Localizing Speech Deepfakes Beyond Transitions via Segment-Aware Learning
by: Mao, Yuchen, et al.
Published: (2026)

MagiCodec: Simple Masked Gaussian-Injected Codec for High-Fidelity Reconstruction and Generation
by: Song, Yakun, et al.
Published: (2025)

SpeechComposer: Unifying Multiple Speech Tasks with Prompt Composition
by: Wu, Yihan, et al.
Published: (2024)

Less is More: Data Curation Matters in Scaling Speech Enhancement
by: Li, Chenda, et al.
Published: (2025)

From Sharpness to Better Generalization for Speech Deepfake Detection
by: Huang, Wen, et al.
Published: (2025)

SpeechFake: A Large-Scale Multilingual Speech Deepfake Dataset Incorporating Cutting-Edge Generation Methods
by: Huang, Wen, et al.
Published: (2025)

Lightweight Front-end Enhancement for Robust ASR via Frame Resampling and Sub-Band Pruning
by: Zhao, Siyi, et al.
Published: (2025)

URGENT Challenge: Universality, Robustness, and Generalizability For Speech Enhancement
by: Zhang, Wangyou, et al.
Published: (2024)

VoiceShop: A Unified Speech-to-Speech Framework for Identity-Preserving Zero-Shot Voice Editing
by: Anastassiou, Philip, et al.
Published: (2024)

An Explicit Consistency-Preserving Loss Function for Phase Reconstruction and Speech Enhancement
by: Ku, Pin-Jui, et al.
Published: (2024)

A Data-Centric Approach to Generalizable Speech Deepfake Detection
by: Huang, Wen, et al.
Published: (2025)

URGENT-PK: Perceptually-Aligned Ranking Model Designed for Speech Enhancement Competition
by: Wang, Jiahe, et al.
Published: (2025)

Diffusion-based Generative Modeling with Discriminative Guidance for Streamable Speech Enhancement
by: Li, Chenda, et al.
Published: (2024)

SLIDE: Integrating Speech Language Model with LLM for Spontaneous Spoken Dialogue Generation
by: Lu, Haitian, et al.
Published: (2025)

DualSpeechLM: Towards Unified Speech Understanding and Generation via Dual Speech Token Modeling with Large Language Models
by: Wang, Yuanyuan, et al.
Published: (2025)

IntMeanFlow: Few-step Speech Generation with Integral Velocity Distillation
by: Wang, Wei, et al.
Published: (2025)

Lessons Learned from the URGENT 2024 Speech Enhancement Challenge
by: Zhang, Wangyou, et al.
Published: (2025)

Generating Speakers by Prompting Listener Impressions for Pre-trained Multi-Speaker Text-to-Speech Systems
by: Chen, Zhengyang, et al.
Published: (2024)

Data-Efficient Low-Complexity Acoustic Scene Classification via Distilling and Progressive Pruning
by: Han, Bing, et al.
Published: (2024)

PURE Codec: Progressive Unfolding of Residual Entropy for Speech Codec Learning
by: Shi, Jiatong, et al.
Published: (2025)

UniVocal: Unified Speech-Singing Code-Switching Synthesis
by: Shi, Yufei, et al.
Published: (2026)

DeepASMR: LLM-Based Zero-Shot ASMR Speech Generation for Anyone of Any Voice
by: Zhang, Leying, et al.
Published: (2026)

CapTalk: Unified Voice Design for Single-Utterance and Dialogue Speech Generation
by: Su, Xiaosu, et al.
Published: (2026)

UniSS: Unified Expressive Speech-to-Speech Translation with Your Voice
by: Cheng, Sitong, et al.
Published: (2025)