Natural environment sound recognition method based on time-frequency domain statistical feature extraction

Abstract

The invention discloses a natural environment recognition method based on time-frequency domain statistical feature extraction. The method comprises the steps of (1) collecting various types of natural environmental sounds such as the sound of an engine, a car horn and wind noise, and establishing a sound sample library, (2) carrying out window framing processing on a sound sample signal, (3) extracting the statistical features of all frame signal in a time domain, (4) marking the type of a sound source to which feature vectors belong, and establishing a sample feature library, (5) training a feature vector by using a support vector machine, and establishing a training model, (6) extracting the feature vectors of a target sound, (7) carrying out matching classification on the feature vectors of the target sound by using the support vector machine, and (8) providing an recognition result. According to the method, the disadvantages of traditional sound LPCC and MFCC feature extraction methods in the aspect of time and frequency combination are made up, and the type to which various types of sounds belong can be judged.

Description

technical field [0001] The invention belongs to the technical field of sound signal recognition, in particular to a natural environment sound recognition method based on time-frequency domain statistical feature extraction. Background technique [0002] In recent years, the recognition of natural environment sounds has gained widespread attention. The natural environment is full of various sounds, such as the sound of engines and car horns in driving vehicles, construction sounds on construction sites, human voices, and birds and insects. , wind and rain, etc. The recognition of natural environment sounds is an important part of machine monitoring, and it also plays an important role in building smart cities and developing smart homes. [0003] In the current natural environment sound recognition technology, most of the technologies used in feature extraction are borrowed from speech recognition algorithms, including: linear predictive cepstral coefficient (LPCC), Mel frequ...

AI Technical Summary

Problems solved by technology

At the same time, due to the attenuation effect of the sound signal during propagation, a single time-domain feature such as LPCC, ZCR or a single frequency-domain feature such as MFCC cannot accurately describe the natural environment sound signals at different distances

Therefore, the feature extraction method of speech recognition cannot be fully applied in the recognition of natural environment sounds.

Images