An Acoustic Analysis and Comparison of Two Unique and Almost Identical Arabic Emphatic Phonemes

Altamrah, Abdulaziz Saud; Altaheri, Hamdi; Alsuliman, Mohammed; Shehadeh, Osama F.; Alotaibi, Yousef Ajami

doi:10.1109/ems.2017.25

Cited by 1 publication

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The audio spectrum is used directly instead of LPCC or MFCC in some studies for telephone identification, such as Sketches of Spectral Features (SSFs) [26], Random Spectral Features (RSFs) [27], and Labeled Spectral Features (LSFs) [28]. The signal spectrogram is another representation of audio signals that is used in many studies ranging from phoneme identification [29] to speech and speaker recognition [30]. The spectrogram of the audio signal was used in [25] for environmental sound classification.…”

Section: Literature Reviewmentioning

confidence: 99%

Digital Audio Forensics: Microphone and Environment Classification Using Deep Learning

et al. 2021

Self Cite

View full text Add to dashboard Cite

The recording device along with the acoustic environment plays a major role in digital audio forensics. We propose an acoustic source identification system in this paper, which includes identifying both the recording device and the environment in which it was recorded. A hybrid Convolutional Neural Network (CNN) with Long Short-Term Memory (LSTM) is used in this study to automatically extract environments and microphone features from the speech sound. In the experiments, we investigated the effect of using the voiced and unvoiced segments of speech on the accuracy of the environment and microphone classification. We also studied the effect of background noise on microphone classification in 3 different environments, i.e., very quiet, quiet, and noisy. The proposed system utilizes a subset of the KSU-DB corpus containing 3 environments, 4 classes of recording devices, 136 speakers (68 males and 68 females), and 3600 recordings of words, sentences, and continuous speech. This research combines the advantages of both CNN and RNN (in particular bidirectional LSTM) models, called CRNN. The speech signals were represented as a spectrogram and were fed to the CRNN model as 2D images. The proposed method achieved accuracies of 98% and 98.57% for environment and microphone classification, respectively, using unvoiced speech segments.

show abstract