A New Text-Independent Speaker Identification Using Vector Quantization and Multi-layer Perceptron

Biometrics is a technology for physical analysis and human behavior used in authentication. One of the behavioral characteristics associated with a person is sound. A person's voice can be identified based on the person's voice signal characteristics. There are several methods in recognizing speaker sound, such as with Mel Frequency Cepstrum Calculation (MFCC) and Subband Based Cepstral (SBC). This study looked for the effectiveness of the use of MFCC and SBC feature extraction with LBG Vector Quantization matching characteristics. Effective feature extraction methods will be tested for realtime speaker identification. The results obtained from this research is the value of MFCC 32 coefficient more effective than the SBC accurately and the speed of the process of identification of both text-dependent and text-independent speaker. The test results of speaker identification in realtime using MFCC is still not satisfactory because the accuracy of recognition is still below 70%.

Download Full-text

Voice Identification Using MFCC and Vector Quantization

Baghdad Science Journal ◽

10.21123/bsj.2020.17.3(suppl.).1019 ◽

2020 ◽

Vol 17 (3(Suppl.)) ◽

pp. 1019

Author(s):

Bassel Alkhatib ◽

Mohammad Madian Waleed Kamal Eddin

Keyword(s):

Vector Quantization ◽

Speech Processing ◽

Large Scale ◽

Feature Matching ◽

Speaker Identification ◽

Voice Recognition ◽

Security Systems ◽

Voice Identification ◽

New Methodologies ◽

Quick Search

The speaker identification is one of the fundamental problems in speech processing and voice modeling. The speaker identification applications include authentication in critical security systems and the accuracy of the selection. Large-scale voice recognition applications are a major challenge. Quick search in the speaker database requires fast, modern techniques and relies on artificial intelligence to achieve the desired results from the system. Many efforts are made to achieve this through the establishment of variable-based systems and the development of new methodologies for speaker identification. Speaker identification is the process of recognizing who is speaking using the characteristics extracted from the speech's waves like pitch, tone, and frequency. The speaker's models are created and saved in the system environment and used to verify the identity required by people accessing the systems, which allows access to various services that are controlled by voice, speaker identification involves two main parts: the first part is the feature extraction and the second part is the feature matching.

Download Full-text

Hybridization process for text-independent speaker identification based on vector quantization model

2016 IEEE International Conference on Signal and Image Processing (ICSIP) ◽

10.1109/siprocess.2016.7888332 ◽

2016 ◽

Author(s):

Mohammed Djeghader ◽

Qin Huang

Keyword(s):

Vector Quantization ◽

Speaker Identification

Download Full-text

Neural Classifiers for Automated Visual Inspection

Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering ◽

10.1243/pime_proc_1994_208_166_02 ◽

1994 ◽

Vol 208 (2) ◽

pp. 83-89 ◽

Cited By ~ 14

Author(s):

D T Pham ◽

E J Bayro-Corrochano

Keyword(s):

Vector Quantization ◽

Visual Inspection ◽

Vision System ◽

Back Propagation ◽

Learning Vector Quantization ◽

Multi Layer Perceptron ◽

Automated Visual Inspection ◽

Valve Stem

This paper discusses the application of a back-propagation multi-layer perceptron and a learning vector quantization network to the classification of defects in valve stem seals for car engines. Both networks were trained with vectors containing descriptive attributes of known flaws. These attribute vectors (‘signatures’) were extracted from images of the seals captured by an industrial vision system. The paper describes the hardware and techniques used and the results obtained.

Download Full-text

DISCRIMINATIVE NONLINEAR DIMENSIONALITY REDUCTION FOR IMPROVED CLASSIFICATION

International Journal of Neural Systems ◽

10.1142/s0129065794000311 ◽

1994 ◽

Vol 05 (04) ◽

pp. 313-333 ◽

Cited By ~ 2

Author(s):

MARK DOLSON

Keyword(s):

Dimensionality Reduction ◽

Vector Quantization ◽

Nonlinear Dimensionality Reduction ◽

Multi Layer Perceptron ◽

Subspace Methods ◽

Improve Performance ◽

Indirect Approach ◽

Classification Tasks ◽

Mlp Neural Networks

Multi-Layer Perceptron (MLP) neural networks have been used extensively for classification tasks. Typically, the MLP network is trained explicitly to produce the correct classification as its output. For speech recognition, however, several investigators have recently experimented with an indirect approach: a unique MLP predictive network is trained for each class of data, and classification is accomplished by determining which predictive network serves as the best model for samples of unknown speech. Results from this approach have been mixed. In this report, we compare the direct and indirect approaches to classification from a more fundamental perspective. We show how recent advances in nonlinear dimensionality reduction can be incorporated into the indirect approach, and we show how the two approaches can be integrated in a novel MLP framework. We further show how these new MLP networks can be usefully viewed as generalizations of Learning Vector Quantization (LVQ) and of subspace methods of pattern recognition. Lastly, we show that applying these ideas to the classification of temporal trajectories can substantially improve performance on simple tasks.

Download Full-text

Automatic Speaker Identification by Voice Based on Vector Quantization Method

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.j9523.0881019 ◽

2019 ◽

Vol 8 (10) ◽

pp. 2443-2445

Keyword(s):

Feature Extraction ◽

Vector Quantization ◽

Speaker Identification ◽

Voice Recognition ◽

Recognition System ◽

Automatic Identification ◽

Quantization Method ◽

Feature Vectors ◽

Cepstral Features ◽

Code Book

In this paper, the systems of speaker identification of a text-dependent and independent nature were considered. Feature extraction was performed using chalk-frequency cepstral coefficients (MFCC). The vector quantization method for the automatic identification of a person by voice has been investigated. Using the extracted features, the code book from each speaker was built by clustering the feature vectors. Speakers were modeled using vector quantization (VQ). Using the extracted features, the code book from each speaker was built by clustering the feature vectors. Codebooks of all announcers were collected in the database. From the results, it can be said that vector quantization using cepstral features produces good results for creating a voice recognition system.

Download Full-text