Method for generating sound encoder for sound event detection

Abstract

The invention provides a method for generating a sound encoder for sound event detection, which comprises the following steps of: performing distortion processing on unlabeled pre-training audio signals in a first training set to obtain distortion training signals; inputting the distortion training signal into an initial sound encoder to obtain a first feature vector; determining a second feature vector based on the pre-training audio signal and the perceptron set; modifying parameters of the initial sound encoder based on the first feature vector and the second feature vector to obtain a candidate sound encoder; and training the candidate sound encoder through the fine tuning audio signal with the label in the second training set to obtain a target sound encoder. According to the invention, the initial sound encoder is pre-trained through the pre-training audio signal without the label to obtain the candidate sound encoder, and then the candidate sound encoder is fine-tuned through the fine-tuning audio signal with the label, so that the dependence on a strong label sample in the training process is reduced, and the robustness of the sound encoder is improved through distortion processing.

Description

technical field [0001] The present application relates to the field of sound event detection, in particular to a method for generating a sound coder for sound event detection. Background technique [0002] Sound carries a large amount of information about physical time in the daily environment. It can perceive the environment through sound, such as streets, offices, etc., and can also identify individual sound sources, such as car engine sound, footsteps, etc. The method of automatically extracting sound event information has great application potential in urban security, for example, using sound event information to identify activities in the environment, using sound event information to alert sensitive events, and constructing urban sound systems based on sound event information within the city. Spectrum map, search surveillance video based on sound event information, etc. [0003] The sound event can be determined through the sound event detection task (Sound Event Detec...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is that the generation of the existing sound encoder for sound event detection requires a large amount of real audio training data with strong labels, and the lack of real audio training data with strong labels will lead to unsatisfactory detection results

Images