Tone quality abnormal data detection method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a tone quality abnormal data detection method and a device, electronic equipment and a storage medium, and the method comprises the steps: receiving user voicedata, and determining voiced segment voice data and voiceless segment voice data in the user voice data; extracting voiced segment acoustic features in the voiced segment voice data, and extracting voiceless segment acoustic features in the voiceless segment voice data; inputting the voiced segment acoustic features into the trained voiced segment generation model to obtain a voiced segment detection result of the user voice data, and inputting the voiceless segment acoustic features into the trained voiceless segment generation model to obtain a voiceless segment detection result of the uservoice data; wherein the voiced segment generation model is trained by adopting normal voiced segment samples, and the silent segment generation model is trained by adopting normal silent segment samples; and determining whether the user voice data are tone quality abnormal data according to the voiced segment detection result and the voiceless segment detection result. According to the method ofthe invention, the accuracy of tone quality abnormal data detection can be improved.

Description

technical field [0001] The present application relates to the technical field of machine learning, in particular to a method, device, electronic equipment and storage medium for detecting data with abnormal sound quality. Background technique [0002] In the voice evaluation system, if the evaluation equipment breaks down, the evaluation result of the voice evaluation system is unsatisfactory. Therefore, there is a need for a data identification system with abnormal sound quality, which can identify whether the voice data has abnormal sound quality. [0003] The current process of abnormal sound quality data detection is as follows: in the training phase, positive example data (normal sound quality data) and negative example data (abnormal sound quality data) are prepared in advance, and the classification model is trained using positive example data and negative example data; in the testing phase, Extract the corresponding features from the speech data to be tested, and us...

AI Technical Summary

Problems solved by technology

However, in practical applications, abnormal data is ever-changing, and it is difficult to list them completely, which leads to error-prone judgments on abnormal data.

Images