阵列配置-不可知性个性化语音增强使用长期-短期空间连贯性
Yicheng Hsu1, Yonghan Lee1, Mingsian R Bai1
1Department of Power Mechanical Engineering, National Tsing Hua University, No. 101, Section 2, Kuang-Fu Road, Hsinchu 30044, Taiwan.
本研究介绍了一种全新的多道个性化语音增强 (PSE) 系统. 它使用新的空间特征有效地抑制背景噪声,在不需要麦克风阵列细节的情况下实现卓越的性能.
科学领域:
- 信号处理 信号处理
- 声学 声学 在声学方面
- 机器学习 机器学习
背景情况:
- 个性化语音增强 (PSE) 旨在抑制干扰语音.
- 多通道PSE系统在杂的环境中比单通道方法具有优势.
- 为各种家庭麦克风阵列实施多通道PSE是一项挑战.
研究的目的:
- 开发一个数组配置不可知的多通道个性化语音增强 (PSE) 系统.
- 引入一种新的空间特征,用于监测目标音箱活动.
- 为了评估系统在不利的声学条件下的性能.
主要方法:
- 定义了一种新的空间特征,即具有动态遗忘因子的长短空间连贯性 (LSTSC).
- 使用卷积循环网络,并将LSTSC功能作为输入.
- 建议简化LSTSC功能,以降低计算成本.
- 实验将拟议的PSE系统与基线进行比较,使用未见的房间响应和阵列配置.
主要成果:
- 建议使用LSTSC训练的多通道PSE网络实现了卓越的语音增强.
- 该系统在不需要精确了解阵列配置或房间响应的情况下证明了有效性.
- 完整的和简化的LSTSC版本都显示出显著的性能改进.
结论:
- 开发的LSTSC功能可以实现数组配置无关的多通道PSE.
- 拟议的系统为复杂的声学环境中语音增强提供了强大的解决方案.
- 这种方法在现实场景中推进了多道PSE的实际应用.
更多相关视频
05:48Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
Published on: August 9, 2024
08:25Combined Invasive Subcortical and Non-invasive Surface Neurophysiological Recordings for the Assessment of Cognitive and Emotional Functions in Humans
Published on: May 19, 2016
相关概念视频
Perceiving Loudness, Pitch, and Location
Place theory, or place coding, suggests that different pitches are heard because various sound waves activate specific locations along the cochlea's basilar membrane. The brain determines the pitch of a sound by...
Linear Approximation in Frequency Domain
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
Perception of Sound Waves
The pitch of a sound depends on the frequency and the pressure amplitude of the source. Two sounds of the same...
¹³C NMR: Distortionless Enhancement by Polarization Transfer (DEPT)
Chunking and Rehearsal in Sensory Memory
Reconstruction of Signal using Interpolation
