Speaker Recognition System : Learn about the Simple and Hybrid Source code for Speaker Identification, based on Neural Networks. Speaker verification is a process through which a machine authenticates the claimed identity of a person from the characteristics of his/her voice. Such systems can be used for providing security in the areas where telephone-mediated transactions are performed. This algorithm provides some form of anatomical or biometric identification i.e. it cannot be lost, forgotten or stolen. Since Artificial Neural Network (ANN) 5 shows a great potentiality in the ﬁeld of speech recognition, we use ANN for evaluating the performance of a variant with multi-Layer perceptron for the task of speaker verification. To prove the effectiveness of the algorithm, the test is carried out with the classification of 2 speakers using a single assertion. The Clustering algorithm partitions the input and output synaptic weights of the trained networks in accordance with a Euclidean distance measure. With this experiment, it is found that the input synaptic weights appear to characterize the speaker in an effective manner. The outcome of the test shows that the chosen ANN model can be used for speaker identiﬁcation and verification purposes.
Angela Wang, Student
I downloaded number of free matlab projects provided by matlabprojects.co.uk for different purpose and all of them are innovative and surprisingly easy to implement. The code flows smoothly and are simple, easy to understand.I liked this project(Speaker Recognition System) in particular due to it’s fully customizable feature.