基于扩散模型的目标语音提取集成处理方法OA
Integrated Processing Method for Target Speech Extraction Based on Diffusion Model
为提升单通道目标语音提取的性能,本文提出了一种基于深度学习的两阶段处理方法.该方法旨在解决现有判别模型过度优化客观指标而导致输出语音产生伪影、失真,进而损害主观听感的问题.本文创新性地引入扩散模型作为后处理优化模块,该模块基于随机微分方程(Stochastic differential equation,SDE)框架,对判别模型的初步输入进行再生式优化.这样不仅有效降低了错误提取,更显著增强了目标语音的自然度与谐波结构的清晰度.实验结果表明,在WSJ0-2mix-extr数据集上,所提方法在模拟主观听觉感知的非侵入式语音质量评估(Non-intrusive speech quality assessment,NISQA)指标上实现了11.18%的提升(从3.22增至3.58),显著改善了语音的主观质量与自然度.主观听测结果比较平均意见得分(Comparative mean opinion score,CMOS)进一步证实了该方法在提升语音清晰度与可懂度方面的有效性.
This paper introduces a two-stage deep learning approach to enhance the performance of single-channel target speech extraction.Conventional discriminative models often over-optimize objective metrics,leading to unnatural artifacts,distortions,and compromised subjective auditory quality in the extracted speech.To address this limitation,we innovatively incorporate a diffusion model as a second-stage refinement module.This module,built upon a stochastic differential equation(SDE)framework,regenerates and refines the initial output from the discriminative model.This process not only reduces extraction errors but also significantly improves the naturalness and harmonic clarity of the target speech.Experimental results on the WSJ0-2mix-extr dataset show that our method achieves an 11.18%improvement(from 3.22 to 3.58)on the non-intrusive speech quality assessment(NISQA)metric,which simulates subjective auditory perception.This indicates a substantial enhancement in perceived speech quality and naturalness.Furthermore,subjective listening tests using the comparative mean opinion score(CMOS)confirm the effectiveness of our approach in boosting speech clarity and intelligibility.
王刚;牟昭曦;贾春波;周立新;顾保国
太原卫星发射中心,太原 030027西安交通大学计算机学院,西安 710072国防科技大学计算机学院,长沙 410073太原卫星发射中心,太原 030027太原卫星发射中心,太原 030027
信息技术与安全科学
语音分离扩散模型语音提取说话人信息生成模型
speech separationdiffusion modelspeech extractionspeaker informationgenerative model
《数据采集与处理》 2026 (5)
1378-1394,17
评论