Abnormal sound monitoring method based on large-scale farm mammals

Abstract

The invention discloses an abnormal sound monitoring method based on large-scale farm mammals, belongs to the field of sound recognition, and particularly relates to an unsupervised sound recognitionmethod. The method mainly comprises the following parts: 1. spectrogram analysis: analyzing collected voice frequencies to determine feasibility of a voice recognition scheme; 2. voice frequency noisereduction: performing noise reduction processing on the voice frequencies to improve accuracy of voice recognition; 3. unsupervised voice frequency segmentation: simplifying a voice frequency processing process without manual segmentation to obtain voice frequency segments containing required sound events; 4. voice frequency feature extraction: adopting a Mel frequency cepstrum coefficient as a feature extraction technology; and 5. unsupervised classification: adopting an unsupervised classification method as a K-means algorithm. According to the method provided by the invention, unsupervisedsound recognition of large-scale farm animals is realized by adopting an unsupervised voice frequency segmentation technology and a K-means classification method, combining a spectrum and time spectrum analysis technology, a voice frequency noise reduction technology and a Mel frequency cepstrum coefficient feature extraction technology.

Description

technical field [0001] The invention belongs to the field of voice recognition, and in particular relates to an unsupervised voice recognition method. Background technique [0002] Voice recognition technology is widely used and has been studied in various fields such as public security, medical care, and intelligent farming. In the existing technologies, the sound recognition technology mostly adopts the supervised learning method, which requires manual participation in audio segmentation and labeling. The process of sound processing and recognition is complicated and the cost is high. In 2015, Fuzhou University invented an animal sound recognition method based on dual features of spectrogram (CN104882144A). By establishing a sound sample library, the pre-stored sound samples and the sound signals to be recognized are converted into spectrograms, and the spectrograms are processed. Normalize, and carry out eigenvalue decomposition and projection, use the double-layer featu...

AI Technical Summary

Problems solved by technology

This method needs to use supervised learning to set the threshold according to the energy of abnormal frames. The threshold setting has a great impact on the classification accuracy, and the operation is relatively complicated, which cannot achieve the purpose of simple and reliable identification.

Images