A robust speaker recognition method based on spectrogram denoising and adversarial learning
Through the joint training of U-Net and TDNN-CGAN, the performance degradation of the speaker recognition system in the noisy environment is solved, and a higher recognition accuracy is achieved.
CN116469394BActive Publication Date: 2025-09-02NANCHANG UNIV
2 Cites 0 Cited by
Patent Information
- Application Number
- CN202310425824.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-04-20
- Publication Date
- 2025-09-02
- Estimated Expiration
- 2043-04-20
AI Technical Summary
Technical Problem
The performance of existing speaker recognition systems in noisy environments has deteriorated, making it difficult to effectively extract clear speech features, resulting in a decrease in recognition accuracy.
Method used
U-Net Mel spectrogram enhancement network and the condition-generated adversarial network TDNN-CGAN based on delay neural network are used. Combined with the speaker classifier, deep features are extracted from noisy speech through spectral denoising and adversarial learning, and close to clean speech features are trained.
Benefits of technology
It improves the recognition accuracy of the speaker recognition system in a noisy environment and achieves higher robustness.
✦ Generated by Eureka AI based on patent content.
Abstract
The present invention provides a robust speaker recognition method based on spectrogram denoising and adversarial learning. First, a spectrogram dataset of clean speech and a noisy spectrogram dataset after the clean speech is noisy are collected; a U-net with a multi-level encoding and decoding structure is trained using a mean square error loss function to remove noise interference from the mel-spectrogram of the noisy speech signal to obtain an enhanced mel-spectrogram; a conditional generative adversarial network based on a time-delay neural network (TDNN-CGAN) is trained using a least squares loss function, a time-delay neural network (TDNN) is used as a generator in the TDNN-CGAN to extract deep features of the enhanced mel-spectrogram, and a multi-layer perceptron (MLP) is used as a discriminator in the TDNN-CGAN; finally, a speaker classifier is trained using cross-entropy loss to identify the speaker's identity, thereby realizing speaker recognition in a noisy environment. The deep features extracted from the noisy speech by the present invention are close to the deep features extracted from the clean speech, thereby improving the performance of the speaker recognition system in a noisy environment.
Need to check novelty before this filing date? Find Prior Art