Sound event recognition method based on optimized parallel model combination

Abstract

The invention relates to a sound event recognition method based on optimized parallel model combination. The sound event recognition method includes 1) recording data of a sound event, acquiring a GMM (Gaussian mixture model) according to clean sound event training, and establishing a clean sound event template; 2) acquiring noise data in current environment in indoor actual noisy environment, acquiring a GMM according to the noise data training, and establishing a noise template; 3) processing the noise template and the clean sound event template by the method of optimized parallel model combination, and obtaining a template of a sound event with noise; 4) sampling to obtain sample signals of the sound event with noise, recognizing sound of the sample signals according to the parameters in the template of the sound event with noise. According to the sound event recognition method, a GMM capable of describing background noise feature distribution better is established as one input in a PMC (portable media center) method, a clean GMM of five sound events is established at another input in the PMC. Meanwhile, robustness of a recognition system to noises is guaranteed.

Description

technical field [0001] The invention belongs to the field of audio signal processing in intelligent monitoring, and relates to a sound event recognition method in an indoor environment, in particular to a sound event recognition method based on an improved parallel model combination. Background technique [0002] Compared with the mature speech recognition methods in the field of artificial intelligence, using computers to recognize sound events is a relatively new research direction in recent years. Sound event recognition is aimed at automatic judgment and classification of sound events that have a certain meaning or can reflect people's behavior in the natural environment. In the home intelligent monitoring system, the recognition of sound events can help people remotely monitor what happens in the home indoor environment, and inform users of what kind of events have occurred in time, which is beneficial for users to deal with them in a timely manner. However, there are ...

AI Technical Summary

Problems solved by technology

[0009] Most of the existing methods still use a set of speech recognition, and the noise processing is no more than the above methods. Among the above methods, the PMC-based method can fully describe the environmental noise and is widely used. They can fully mine the environmental noise. information to improve the robustness of system identification, but the existing PMC method uses a single Gaussian model (SGM) to describe the noise characteristics. For the complex noise, SGM cannot characterize the characteristics of the noise well.

Therefore, the recognition rate is not ideal in the case of complex noise

Images