Source microphone identification from speech recordings based on a Gaussian mixture model

Eskidere, Ömer

doi:10.3906/elk-1207-74

Cited by 22 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this approach, voice samples were modeled as a weighted sum of multivariate Gaussian probability density functions. In the GMM parameter estimation, the distribution of features is modeled by the mean vectors

, covariance matrices ∑ i , and mixture weights c i which is denoted by the notation Θ = { c i , μ i , ∑ i }, i = 1,2,…, K , where K is the number of mixture components [ 39 ]. These model parameters (Θ) are commonly determined using expectation maximization (EM) algorithm.…”

Section: Methodsmentioning

confidence: 99%

Voice Disorder Classification Based on Multitaper Mel Frequency Cepstral Coefficients Features

Eskidere

Gürhanlı

2015

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

The Mel Frequency Cepstral Coefficients (MFCCs) are widely used in order to extract essential information from a voice signal and became a popular feature extractor used in audio processing. However, MFCC features are usually calculated from a single window (taper) characterized by large variance. This study shows investigations on reducing variance for the classification of two different voice qualities (normal voice and disordered voice) using multitaper MFCC features. We also compare their performance by newly proposed windowing techniques and conventional single-taper technique. The results demonstrate that adapted weighted Thomson multitaper method could distinguish between normal voice and disordered voice better than the results done by the conventional single-taper (Hamming window) technique and two newly proposed windowing methods. The multitaper MFCC features may be helpful in identifying voices at risk for a real pathology that has to be proven later.

show abstract

Section: Methodsmentioning

confidence: 99%

Voice Disorder Classification Based on Multitaper Mel Frequency Cepstral Coefficients Features

Eskidere

Gürhanlı

2015

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

show abstract

“…Several studies employed MFCC for microphone and environment classification [7], [19], [24], [25]. Several other frame-level features have also been evaluated for microphone recognition, such as Multi-taper MFCC [9], or Linear Prediction Cepstral Coefficient (LPCC) [10]. The researchers in [10] classified 16 different microphones recorded in one silent room using a Gaussian Mixture Model (GMM) classifier.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Several other frame-level features have also been evaluated for microphone recognition, such as Multi-taper MFCC [9], or Linear Prediction Cepstral Coefficient (LPCC) [10]. The researchers in [10] classified 16 different microphones recorded in one silent room using a Gaussian Mixture Model (GMM) classifier. They studied 3 feature representations: LPCC, MFCC, and Perceptually-based Linear Predictive Coefficients (PLPCs).…”

Section: Literature Reviewmentioning

confidence: 99%

Digital Audio Forensics: Microphone and Environment Classification Using Deep Learning

et al. 2021

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

“…Hanilci et al [18] explored microphone identification problem of speech recording from a mobile phone where audio recorded from 14 models of mobile phones are classified using vector quantization and SVM-based classifier. Moreover, Eskidere [19] reported his recent work in microphone identification on 16 microphone models using GMM-based modeling technique along with three different features called LPCC, PLPC and MFCC.…”

Section: Introductionmentioning

confidence: 99%

Statistical Based Audio Forensic on Identical Microphones

Kurniawan

Rahim

Khan

2016

IJECE

View full text Add to dashboard Cite

<p>Microphone forensics has become a challenging field due to the proliferation of recording devices and explosion in video/audio recording. Video or audio recording helps a criminal investigator to analyze the scene and to collect evidences. In this regards, a robust method is required to assure the originality of some recordings. In this paper, we focus on digital audio forensics and study how to identify the microphone model. Defining microphone model will allow the investigators to conclude integrity of some recordings. We perform statistical analysis on the recording that is collected from two microphones of the same model. Experimental results and analysis indicate that the signal of sound recording of identical microphone is not exactly same and the difference is up to 1% - 3%.</p>

show abstract

Source microphone identification from speech recordings based on a Gaussian mixture model

Cited by 22 publications

References 26 publications

Voice Disorder Classification Based on Multitaper Mel Frequency Cepstral Coefficients Features

Voice Disorder Classification Based on Multitaper Mel Frequency Cepstral Coefficients Features

Digital Audio Forensics: Microphone and Environment Classification Using Deep Learning

Statistical Based Audio Forensic on Identical Microphones

Contact Info

Product

Resources

About