Turkish Journal of Electrical Engineering and Computer Sciences
Abstract
This paper presents a new method for extraction of accent information from Urdu speech signals. Accent is used in speaker recognition system especially in forensic cases and plays a vital role in discriminating people of different groups, communities and origins due to their different speaking styles. The proposed method is based on Gaussian mixture model-universal background model (GMM-UBM), mel-frequency cepstral coefficients (MFCC), and a data augmentation (DA) process. The DA process appends features to base MFCC features and improves the accent extraction and forensic speaker recognition performances of GMM-UBM. Experiments are performed on an Urdu forensic speaker corpus. The experimental results show that the proposed method improves the equal error rate and the accuracy of GMM-UBM by 2.5 % and 3.7 %, respectively.
DOI
10.3906/elk-1812-152
Keywords
Forensic, classification, speaker recognition, speech features
First Page
3763
Last Page
3778
Recommended Citation
TAHIR, F, SALEEM, S, & AHMAD, A (2019). Extracting accent information from Urdu speech for forensic speaker recognition. Turkish Journal of Electrical Engineering and Computer Sciences 27 (5): 3763-3778. https://doi.org/10.3906/elk-1812-152
Included in
Computer Engineering Commons, Computer Sciences Commons, Electrical and Computer Engineering Commons
