A method of train whistle recognition in complex noise environment

A recognition method and noise technology, applied in speech analysis, instruments, etc., can solve problems such as sample redundancy and increased calculation cost of training classifiers, and achieve the effects of overcoming data overflow, scientific and reasonable preprocessing results, and improving classifier performance

Active Publication Date: 2019-04-23
HEFEI UNIV OF TECH
View PDF16 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In the traditional training process, there are the following problems: First, when the number of training samples is large enough, the classification learning method based on statistics can obtain a classifier with strong generalization ability, but the computational cost required to train the classifier will also increase. Then it increases; second, there is a problem of sample redundancy in many sample libraries, and similar samples do not need repeated training; third, in reality, the train sound is a complex sound signal that contains various sound types and various types appear alternately

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method of train whistle recognition in complex noise environment
  • A method of train whistle recognition in complex noise environment
  • A method of train whistle recognition in complex noise environment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] In this embodiment, a train whistle recognition method under complex noise environment, refer to figure 1 , proceed as follows:

[0049] Step 1: Use the microphone to obtain the W when the train passes by 1 original sound samples, denoted as S={S(1),S(2),…,S(m),…,S(W 1 )}, S(m) represents the mth original sound sample; record the time length of the mth original sound sample S(m) as T(m), 1≤m≤W 1 ; In the process of collecting samples, W 1 The larger the value of , the better, so that the training samples can more fully reflect the actual situation. In this example, the W 1 The value of is set to 200, and the time length T(m) ranges from 30 seconds to 180 seconds. The properties of the sound files are all sampling rate 48kHz, 16bit, single channel, format is wav, PCM encoding form.

[0050] Step 2: Refer to figure 2 The process of selecting a representative training sample set;

[0051] Step 2.1, manually identify W 1 Whistle segment and non-whistle segment in ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a train whistle recognizing method for a complex noise environment. The method is characterized in steps of 1, utilizing a microphone for obtaining an original training sample base; 2, selecting a representative training sample set; 3, utilizing an HMM model for training the training sample set and obtaining a model base; 4, utilizing the microphone for obtaining a testing sample base, then utilizing the HMM model for classified recognition of the testing sample base, and obtaining a final recognition result. According to the invention, the training sample set with high quality can be obtained with comparatively less manual marking, so that difficulties in training sample selection caused by train noise complexity are eliminated and the recognition correctness is improved further.

Description

technical field [0001] The invention relates to a train whistle recognition method in a complex noise environment, and belongs to the technical field of sound recognition. Background technique [0002] The sound signal has the advantage of not being affected by light and vision, and its identification and analysis can obtain information that cannot be captured by vision. Therefore, voice recognition is widely used in security, navigation, environmental sound detection and intelligent traffic detection and other fields. In recent years, the research on speech recognition has been quite mature, but the research on non-speech sounds is far behind speech recognition. At present, there is still a lack of systematic methods for the recognition of non-speech sounds, and most studies directly use the feature extraction and classification methods in speech recognition technology. [0003] The commonly used features in the feature extraction process are Mel frequency cepstral coeffi...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G10L25/51G10L25/24
CPCG10L25/24G10L25/51
Inventor 蒋翠清樊鹏丁勇邵宏波
Owner HEFEI UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap