Voice Activity Detection: Merging Source and Filter-based Information

Abstract

Voice Activity Detection (VAD) refers to the problem of distinguishing speech segments from background noise. Numerous approaches have been proposed for this purpose. Some are based on features derived from the power spectral density, others exploit the periodicity of the signal. The goal of this paper is to investigate the joint use of source and filter-based features. Interestingly, a mutual information-based assessment shows superior discrimination power for the source-related features, especially the proposed ones. The features are further the input of an artificial neural network-based classifier trained on a multi-condition database. Two strategies are proposed to merge source and filter information: feature and decision fusion. Our experiments indicate an absolute reduction of 3% of the equal error rate when using decision fusion. The final proposed system is compared to four state-of-the-art methods on 150 minutes of data recorded in real environments. Thanks to the robustness of its source-related features, its multi-condition training and its efficient information fusion, the proposed system yields over the best state-of-the-art VAD a substantial increase of accuracy across all conditions (24% absolute on average).

Effective emotion recognition in movie audio tracks

“…The binary classification problem is described as Regarding the feature extraction part of this paper, it uses a relatively small number of features that are fed as input to an ensemble classification schema. We tested Clarity, that is commonly used in voice activity detection [20], [21] as a candidate feature for emotion recognition. The reason for that choice is that emotion recognition is a complex, versatile task, so alternative features may capture supplementary aspects of emotion expression.…”

A robust polynomial regression-based voice activity detector for speaker verification

“…The speaker models were then adapted from their respective gender's UBM with a relevance factor of 16. For the i-vector framework, same UBMs and the pooled training data were used to train the total variability matrix in 20 iterations, and then, 100 dimensional i-vectors were Table 1 Male speaker verification results of GMM-UBM method in terms of percent EER (minDCF) for the proposed algorithm, Drugman's VAD method [27], and Rangachari's noise tracking method [21]. The last columns show the relative percent EER reduction rates compared to Drugman's VAD and Rangachari's method, respectively extracted from each utterance.…”

“…One of these methods was the noise tracking algorithm proposed in [21] (called Rangachari's method from here on). This algorithm works on the frequency bins of the conventional Table 2 Female speaker verification results of GMM-UBM method in terms of percent EER (minDCF) for the proposed algorithm, Drugman's VAD method [27], and Rangachari's noise tracking method [21]. The last columns show the relative percent EER reduction rates compared to Drugman spectrum, but, to make a fair comparison, it was modified to work on the mel spectrum, similar to the algorithm proposed in this paper.…”

“…Four different voicing measures and a spectral feature are concatenated and mapped to a onedimensional space via principal component analysis in [26], and the resulting feature distribution is bi-Gaussian. In addition to the four voicing measures used in [26], MFCCs and two pitch trackers are used as features for the VAD in [27]. A comparison of some standard VAD methods for speaker recognition can be found in [28], where the bi-Gaussian modeling-based VAD is reported to be the best performing one.…”

“…A polynomial is fitted on the energy output over a range of several frames, and the output is not strictly frame-dependent. Also, suppression of sudden energy ripples is possible with this method, which is achieved by smoothing in some studies [27]. Besides the polynomial regression, another important novelty of the proposed algorithm is the pseudo SNR-dependent linear thresholding that takes account of the speech presence in each band.…”