Informatics Publishing Limited

Syed Abbas Ali ¹, Sallar Khan ², Humaira Perveen ¹, Reham Muzzamil ¹, Mahnoor Malik ¹, Faiza Khalid ¹

Affiliations
1 NED University of Engineering and Engineering and Technology, Main University Road، Karachi 75270, IN
2 Sir Syed University of Engineering and Engineering and Technology, Karachi, PK

Abstract

Urdu language of Pakistan has more than 100 million speakers in Pakistan, India, Afghanistan and Middle East. With low English literacy rate average Urdu speaking person faces barriers in communicating with foreign people in terms of accessing information, carrying business. This paper proposes an interactive Urdu to English language speech translator using deep Neural Network. ASR module in proposed pipeline is composed of deep neural network and is simpler as compared to traditional ASR which requires complex hand engineering like feature extraction and resources like phoneme dictionary. It was clearly seen that the proposed model shows the high accuracy when the input is recorded audio and it shows poor performance with real time input. While one HTTP request per input transcription produced English translation for Text to Text translation using Python Text Blob library. The final output was achieved with a delay of no more than 30 seconds. Furthermore, we have tested and provided some statistical findings, the result shows that value updating for neural network layer’s bias, standard deviation when Adam optimizer parameters are set as follows: beta1=0.9, beta2=0.9 and learning rate =0.01 meanwhile dropout rate was kept to 5% to offer regularization and observed value for scalar maximum lies between 0 and 0.08. There is a little deviation at 0.05 step, value decreases and afterwards that bias maximum scalar increases with positive values and finally increases exponentially at later stages of training further results are discussed in experiment section respectively. The proposed speech recognition model out performs traditional automatic speech recognition systems in efficiency, simplicity and robustness.

Keywords

Deep RNN, Language Translator, N-gram LM, Text Blob Translation, Urdu Language, ASR

Full Text

   

Syed Abbas Ali ¹, Fatima Waheed ¹, Wajahat Rehman ¹, Sallar Khan ¹, Marium Zia ¹, Jawaria Sallar ²

Affiliations
1 NED University of Engineering and Technology, Karachi, Karachi City, Sindh − 75270, PK
2 Sir Syed University of Engineering and Technology, Block 5 Gulshan-e-Iqbal, Karachi, Sindh − 75270, PK

Abstract

Lung Cancer detection making use of medical imaging is still a challenging task for radiologist. The objective of this research is to classify the types of lung tumours for extracted and selected features using learning algorithms. In this paper, an experimental study is conducted on 100 cases of lung cancer to evaluate the performance of learning classifiers (DNN, SVM, Random Forest, Decision Tree, Naïve Bayes) with different medical Imaging (DICOM) features to identify the two types of Lung cancer (Benign and Malignant). The proposed methodology intends to automate the entire procedure of diagnosis by automatically detecting the tumor, measuring the required values such as diameter, perimeter, area, centroid, roundness, indentations and calcification. Experiment is conducted in to two phases: In the first phase, identify the most significant feature used in lung cancer analysis by CT scan and perform the mapping to computer related format. In the second phase, feature selection and extraction is performed to machine learning algorithms. To evaluate the performance of classifiers in term of classification accuracy and improving the false positive rate, every stage of evolution is divided into four different phases: single phase module, single slice testing, series testing and testing of learning algorithms. Experimental results show significant improvement in false positive rate up to 30% for both Benign and Malignant. Whereas, Deep Neural Network (DNN) demonstrate high values in term of classification accuracy in comparison with other classifiers. The proposed methodology for lung cancer detection system having a potential to reduce the time and cost of diagnosis procedure and use for early detection of lung cancer.

Full Text