Zakariya Qawaqneh

Deep neural network framework and transformed MFCCs for speaker's age and gender classification

Abstract Speaker age and gender classification is one of the most challenging problems in speech processing. Although many studies have been carried out focusing on feature extraction and classifier design for improvement, classification accuracies are still not satisfactory. The key issue in identifying speakers age and gender is to generate robust features and to design an in-depth classifier. Age and gender information is concealed in speakers speech, which is liable for many factors such as, background noise, speech contents, and phonetic divergences. The success of DNN architecture in many applications motivated this work to propose a new speakers age and gender classification system that uses BNF extractor together with DNN. This work has two major contributions: Introduction of shared class labels among misclassified classes to regularize the weights in DNN and generation of transformed MFCCs feature set. The proposed system uses HTK to find tied-state triphones for all utterances, which are used as labels for the output layer in the DNNs for the first time in age and gender classification. BNF extractor is used to generate transformed MFCCs features. The performance evaluation of the new features is done by two classifiers, DNN and I-Vector. It is observed that the transformed MFCCs are more effective than the traditional MFCCs in speakers age and gender classification. By using the transformed MFCCs, the overall classification accuracies are improved by about 13%.

Age and gender classification from speech and face images by jointly fine-tuned deep neural networks

A new cost function is developed for the concurrent and jointly fine-tuned DNNs.A new DNN architectures are proposed.Two DNNs are trained concurrently and jointly fine-tuned.Speaker age and gender classification is studied.Age estimation from unconstrained face images is studied. The classification of humans age and gender from speech and face images is a challenging task that has important applications in real-life and its applications are expected to grow more in the future. Deep neural networks (DNNs) and Convolutional neural networks (CNNs) are considered as one of the state-of-art systems as feature extractors and classifiers and are proven to be very efficient in analyzing problems with complex feature space. In this work, we propose a new cost function for fine-tuning two DNNs jointly. The proposed cost function is evaluated by using speech utterances and unconstrained face images for age and gender classification task. The proposed classifier design consists of two DNNs trained on different feature sets, which are extracted from the same input data. Mel-frequency cepstral coefficients (MFCCs) and fundamental frequency (F0) and the shifted delta cepstral coefficients (SDC) are extracted from speech as the first and second feature sets, respectively. Facial appearance and the depth information are extracted from face images as the first and second feature sets, respectively. Jointly training of two DNNs with the proposed cost function improved the classification accuracies and minimized the over-fitting effect for both speech-based and image-based systems. Extensive experiments have been conducted to evaluate the performance and the accuracy of the proposed work. Two publicly available databases, the Age-Annotated Database of the German Telephone Speech database (aGender) and the Adience database, are used to evaluate the proposed system. The overall accuracy of the proposed system is calculated as 56.06% for seven speaker classes and overall exact accuracy is calculated as 63.78% for Adience database.

New transformed features generated by deep bottleneck extractor and a GMM–UBM classifier for speaker age and gender classification

Speaker age and gender classification is one of the most challenging problems in speech signal processing. Recently with developing technologies, identifying speaker age and gender information has become a necessity for speaker verification and identification systems such as identifying suspects in criminal cases, improving human–machine interaction, and adapting music for awaiting people queue. Despite the intensive studies that have been conducted to extract descriptive and distinctive features, the classification accuracies are still not satisfactory. In this work, a model for generating bottleneck features from a deep neural network and a Gaussian Mixture Model–Universal Background Model (GMM–UBM) classifier are proposed for speaker age and gender classification problem. Deep neural network with a bottleneck layer is trained in an unsupervised manner for calculating the initial weights between layers. Then, it is trained and tuned in a supervised manner to generate transformed mel-frequency cepstral coefficients (T-MFCCs). The GMM–UBM is used to build a GMM model for each class, and the models are used to classify speaker age and gender. Age-annotated database of German telephone speech (aGender) is used to evaluate the proposed classification system. The newly generated T-MFCCs have shown potential to achieve significant classification improvements in speaker age and gender classification by using the GMM–UBM classifier. The proposed classification system achieved an overall accuracy of 57.63%. The highest accuracy is calculated as 72.97% for adult female speakers.

A new hardware quantum-based encryption algorithm

Cryptography is entering a new age since the first steps that have been made towards quantum computing, which also poses a threat to the classical cryptosystem in general. In this paper, we introduce a new novel encryption technique and algorithm to improve quantum cryptography. The aim of the suggested scheme is to generate a digital signature in quantum computing. An arbitrated digital signature is introduced instead of the directed digital signature to avoid the denial of sending the message from the sender and pretending that the senders private key was stolen or lost and the signature has been forged. The onetime pad operation that most quantum cryptography algorithms that have been proposed in the past is avoided to decrease the possibility of the channel eavesdropping. The presented algorithm in this paper uses quantum gates to do the encryption and decryption processes. In addition, new quantum gates are introduced, analyzed, and investigated in the encryption and decryption processes. The authors believe the gates that are used in the proposed algorithm improve the security for both classical and quantum computing. (Against)The proposed gates in the paper have plausible properties that position them as suitable candidates for encryption and decryption processes in quantum cryptography. To demonstrate the security features of the algorithm, it was simulated using MATLAB simulator, in particular through the Quack Quantum Library.

Combining Two Different DNN Architectures for Classifying Speaker's Age and Gender.

Speakers’ age and gender classification is one of the most challenging problems in the field of speech processing. Recently, remarkable developments have been achieved in the neural network field, nowadays, deep neural network (DNN) is considered one of the state-of-art classifiers which have been successful in many speech applications. Motivated by DNN success, we jointly fine-tune two different DNNs to classify the speaker’s age and gender. The first DNN is trained to classify the speaker gender, while the second DNN is trained to classify the age of the speaker. Then, the two pre-trained DNNs are reused to tune a third DNN (AGender-Tuning) which can classify the age and gender of the speaker together. The results show an improvement in term of accuracy for the proposed work compared with the I-Vector and the GMM-UBM as baseline systems. Also, the performance of the proposed work is compared with other published works on a publicly available database.

DNN-based Models for Speaker Age and Gender Classification.

Automatic speaker age and gender classification is an active research field due to the continuous and rapid development of applications related to humans’ life and health. In this paper, we propose a new method for speaker age and gender classification, which utilizes deep neural networks (DNNs) as feature extractor and classifier. The proposed method creates a model for each speaker. For each test speech utterance, the similarity between the test model and the speaker class models are compared. Two feature sets have been used: Melfrequency cepstral coefficients (MFCCs) and shifted delta cepstral (SDC) coefficients. The proposed model by using the SDC feature set achieved better classification results than that of MFCCs. The experimental results showed that the proposed SDC speaker model + SDC class model outperformed all the other systems by achieving 57.21% overall classification accuracy.