Research and application of voiceprint recognition based on a deep recurrent neural network

Voiceprint recognition is one of the most popular biometric recognition technologies, which can recognize the identity of the speaker through the voice. Based on Convolutional Neural Network (CNN) and deep Recurrent Neural Network (RNN) voiceprint recognition program, the research of voiceprint recognition technology is called CDRNN. The CDRNN processes the speaker's original speech information through a series of processes and generates a two-dimensional spectrogram. The CNN is better than the advantage of processing the image to extract the personality characteristics of the speech signal from the spectrogram. These personality features are then inputted into the deep RNN. Voiceprint identification is then used to determine the identity of the speaker. The experimental results showed that the CDRNN can obtain better recognition accuracy than other schemes such as Gaussian Mixture Model-Universal Background Model (GMM-UBM).