Audio data classification method for region environment
An ecological environment and audio data technology, which is applied in the fields of electrical digital data processing, special data processing applications, instruments, etc., can solve the problems of retrieval and classification of audio data not involving regional ecological environment, etc., so as to improve the efficiency of classification and be more practical. The effect of stability and broad application prospects
Inactive Publication Date: 2010-07-21
FUZHOU UNIV
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] However, none of these studies involved the retrieval and classification of regional ecological environment audio data
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0032] Regional eco-environmental audio data classification method of the present invention, the method comprises the following steps:
[0033] (1) effectively segment the ecological environment audio signal of the type to be determined, and calculate the segment length of each effective segment, and realize through the following steps:
[0034] a) Framing the audio signal;
[0035] b) Calculate the average amplitude Mi of each frame and the average amplitude M of the entire audio signal;
[0036] c) Taking a section of audio signal whose frame average amplitude Mi is higher than 0.67*M for 5 consecutive frames or more than 5 frames as an effective segment, thereby dividing the audio signal into several effective segments;
[0037] d) Record the number of frames in each valid segment.
[0038] (2) Calculate the difference Mel frequency cepstral coefficient matrix d-mfccm of each effective segment, to obtain the classification parameter of each effective segment therefrom, an...
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
Login to View More Abstract
The invention relates to a technology for searching, classifying and processing audio data and provides a high-efficiency classification method for the analysis, the indexing, the searching and the retrieval of the audio data of a regional entironment. The method comprises the following steps: (1) an audio signal is subjected to effective subsection and the length of each subsection is calculated; 2) a differential Mel frequency cepstrum coefficient matrix of each effective subsection is calculated; 3) the largest harmonic component frequency component of each effective subsection is calculated, and the distribution relation between the length of each subsection and the largest harmonic component frequency component is established; 4) the classification and searching range of each effective subsection is respectively determined respectively, and data points corresponding to the effective subsections of known types of all the audio signals falling in each classification and searching range are searched in the database respectively; 5) a classification parameter of each effective subsection is compared with the classification parameters of all the data points falling in the classification and searching range respectively, and according to the close degree, a possible type of each effective subsection is determined; and 6) according to the possible type of each effective subsection, the type of the audio signal which is to be classified is determined finally.
Description
technical field [0001] The invention relates to audio data retrieval, classification and processing technology, in particular to a regional ecological environment audio data classification method. Background technique [0002] Different regions, locations, years, seasons, climates and different times of day can produce different sounds in the ecological environment. These sounds reflect the survival and activities of various creatures in nature to a certain extent, and are also closely related to the living environment of human beings. Study the classification technology of regional ecological environment audio data, identify, index and retrieve methods and key technologies for ecological environment audio data, determine the relationship between regional ecological environment audio and regions, locations, seasons, climate conditions and time, analyze and study the ecological environment It is of great significance to predict the changes of the ecological environment and u...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F17/30
Inventor 李应
Owner FUZHOU UNIV