Speaker recognition method and system based on gender and language

Abstract

The invention discloses a speaker recognition method and system based on gender and language, and belongs to the field of speaker recognition. Comprising the following steps: acquiring to-be-recognized voice data, specifically an audio file containing an audio of an effective speaker; carrying out noise reduction processing on the audio file to obtain a low-noise voice audio; carrying out SMAC feature extraction on the denoised voice audio to obtain a voice spectrum feature map; inputting the speech spectrum feature map into a ResNet model to obtain a speech feature vector; inputting the voice feature vector into a multi-target learning model, and identifying to obtain the identity of the speaker, the gender of the speaker and the language information used by the speaker; and performing weighted fusion on the three recognition task results to obtain a speaker recognition result corresponding to the to-be-recognized voice data. According to the method, the gender information and the language information in the voice are comprehensively utilized, the robustness of speaking recognition is effectively improved, and particularly, the recognition precision is high under the condition that the voice of the speaker changes.

Description

technical field [0001] The invention relates to the field of speaker recognition, in particular to a gender and language-based speaker recognition method and system. Background technique [0002] With the continuous development of artificial intelligence, more and more intelligent identification technologies have been applied in life, including face recognition, fingerprint recognition and voiceprint recognition that have emerged in recent years. Voiceprint recognition, also known as speaker recognition, analyzes a piece of audio content to identify which speaker the audio belongs to. Speakers can be used for identity authentication and have attracted widespread attention because of their convenience. [0003] In the prior art, most of the speaker recognition methods focus on a single factor, that is, the speaker’s own recognition. This method requires the speaker to keep the same way of speaking in the two stages of voiceprint registration and voiceprint recognition. When ...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a speech recognition method and system based on gender and language, which combines the gender information and language information contained in the voice content to identify the speaker, and solves the problem of a single factor when the tone of the language changes. The technical problem that the accuracy rate of the recognition method drops

Images