Sound Event Detection at Low SNR Based on Multi-random Forests

LI Ying; YIN Jia-li

doi:10.3969/j.issn.0372-2112.2018.11.018

您当前的位置：

首页 >

文章列表页 >

Sound Event Detection at Low SNR Based on Multi-random Forests

更新时间：2025-07-16

- Sound Event Detection at Low SNR Based on Multi-random Forests
- Acta Electronica Sinica Vol. 46, Issue 11, Pages: 2705-2713(2018)
- 作者机构：
  
  1. 福州大学数学与计算机科学学院,福建,福州,350116
  2. 网络系统信息安全福建省高校重点实验室,福建,福州,350116
  3. 福州大学数学与计算机科学学院,福建,福州,350116
  4. 网络系统信息安全福建省高校重点实验室,福建,福州,350116
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China (No.61075022);Natural Science Foundation of Fujian Province (No.2018J01793)
- DOI：10.3969/j.issn.0372-2112.2018.11.018
  CLC： TP391.42
- Published Online：25 November 2018，
  
  Published：2018
- 稿件说明：
移动端阅览
LI Ying, YIN Jia-li. Sound Event Detection at Low SNR Based on Multi-random Forests[J]. Acta Electronica Sinica, 2018, 46(11): 2705-2713.
DOI：

LI Ying, YIN Jia-li. Sound Event Detection at Low SNR Based on Multi-random Forests[J]. Acta Electronica Sinica, 2018, 46(11): 2705-2713. DOI： 10.3969/j.issn.0372-2112.2018.11.018.

摘要

论文针对各种背景声音中低信噪比声音事件的检测问题，提出把背景声音与声音事件混合，形成带噪声样本来训练分类器.在预处理阶段，使用基于经验模态分解与2-6级固有模态函数的投票方法，对背景声音与声音事件端点进行预测并估算信噪比.接着使用子带能量分布方法，提取声音数据的特征.最后，论文将背景声音与声音事件样本库中所有声音样本按照估算的信噪比相混合，生成混合声音特征训练多随机森林，用于低信噪比声音事件的检测.实验证实，所提出的方法可以用于各种声场景下低信噪比声音事件的检测，并能在信噪比为-5dB的情况下保持67.1%的平均检测率.

Abstract

For sound event detection under various background noises at low SNR

this paper proposes a method that mixes the background noises with sound events into noisy samples to train classifiers. In the pre-processing stage

we use a voting method based on 2th to 6th intrinsic mode functions (IMFs) that generated from empirical mode decomposition (EMD)

to detect the endpoint of sound events and estimate the SNR. Then subband power distribution (SPD) is used to extract features from audio data. Finally

we mix the background noise and all the sound event samples in the sound event database according to the estimated SNR

and then extract the noisy samples features to train multi-random forests (M-RF) for the detection of the sound events in low SNR environment. The experiment proves that the proposed method has the ability to recognize sound events in various acoustic scenes at low SNR

and can remain an average accuracy rate of 67.1% at-5dB.

关键词

Keywords

references

Views

235

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Performance Assessment of Coastal Sea Surface Height Measurement Using Opportunity-Source Signal from GEO Satellite

Spatial-Channel Feature Representation and Self-attention Pooling for Weakly-Labeled Sound Event Detection

Local Mean Computation for Empirical Mode Decomposition Based on Higher Order Derivative

Empirical Mode Decomposition De-noising Method Based on Principal Component Analysis

ECG De-noising Based on Hilbert-Huang Transform

Related Author

YANG Dong-kai

WANG Feng

HOU Zhen-wei

HAO Jun-yong

GU Xiao-hua

YANG Li-ping

LI Heng

LI Zhi

Related Institution

School of Electronic Information Engineering， Beihang University

State Key Laboratory of Geo-Information Engineering

School of Electrical Engineering, Chongqing University of Science & Technology

Key Laboratory of Optoelectronic Technology and Systems, Ministry of Education, Chongqing University

School of Electronic Engineering and Automation Guilin University of Electronic Technology Guilin Guangxi China

⁰