Classification Of Singing Voices Research Articles

An Old Myth: Prediction of the Correct Singing Voice Classification. True or not?

<h2>Summary</h2><h3>Objective</h3> Predicting the correct singing voice classification based on laryngoscopy is an old myth. The aim of this study was to evaluate if a professional phoniatrician/laryngologist can predict the correct singing voice classification only from laryngoscopy and to analyze different anatomical parameters between professional sopranos and altos to determine whether a multivariate analysis of anatomical parameters can predict the singing register. <h3>Study design</h3> Prospective study <h3>Methods</h3> We included 49 professional female singers (25 sopranos, 24 altos). Laryngoscopic images were shown to professional phoniatricians/laryngologists to rate whether it they show a soprano or an alto. In addition, a high-resolution computer tomography (HRCT) scan was performed during singing of ƒ0 by each singer. DICOM scan data were rendered and 3D-visualized using the software MIMICS. In all singers, we measured the length of the vocal folds of the glottis, the distance from the anterior commissure orthogonally to the vertebral spine, and the antero-posterior distance of the subglottis/trachea 1 cm and 2 cm below the glottis. We also measured the length and volume of the resonance space. <h3>Results</h3> It was not possible to predict the singing voice classification only from laryngoscopy. In the HRCT images, sopranos had significantly shorter vocal folds, a shorter glottal length, a shorter distance from the anterior commissure to the spine, and a shorter resonance space. When combining all parameters, the chance of correctly predicting a soprano was 74.1% and an alto 68.2%. <h3>Conclusion</h3> Although there are anatomical differences between sopranos and altos, prediction of the singing voice classification from laryngoscopy or HRCT is not reliable enough for clinical use.

Application of Pitch Derived Parameters to Speech and Monophonic Singing Classification

Speech and singing voice discrimination is an important task in the speech processing area given that each type of voice requires different information retrieval and signal processing techniques. This discrimination task is hard even for humans depending on the length of voice segments. In this article, we present an automatic speech and singing voice classification method using pitch parameters derived from musical note information and f 0 stability analysis. We applied our method to a database containing speech and a capella singing and compared the results with other discrimination techniques based on information derived from pitch and spectral envelope. Our method obtains good results discriminating both voice types, is efficient, has good generalisation capabilities and is computationally fast. In the process, we have also created a note detection algorithm with parametric control of the characteristics of the notes it detects. We compared the agreement of this algorithm with a state-of-the-art note detection algorithm and performed an experiment that proves that speech and singing discrimination parameters can represent generic information about the music style of the singing voice.

Type-2 Fuzzy Gaussian mixture models for singing voice classification in commercial music production

The paper describes a system of singing voice classification in the commercial music productions. A first step in our system is to separate the singer's voice from the music. Based on the vocal part, two sets of parameters are formed, one for singing voice type and the other for the singing voice quality. Each set of parameters contains a number of MPEG–7 low–level descriptors and other descriptors; at the classification stage the paper suggests an extension of Gaussian Mixture Models (GMMs), by using the Type–2 FGMMs (Type–2 Fuzzy Gaussian Mixture Models). Results show substantial improvements when compared to similar works.

Classification of the Classical Male Singing Voice Using Long-Term Average Spectrum

Singing-voice classification is often considered the cornerstone of a classical singer's identity. Traditionally, classification has been a highly subjective, nonstandardized process. As a result, misclassification of the singing voice is thought to be common, especially in young singers. Long-term average spectrum (LTAS) average is an objective measurement that could be used to classify a singer's voice. The purpose of this study was to determine the relationship of LTAS with singing-voice classification. Descriptive between-subject study. Nine professional classical male singers performed the "Star Spangled Banner" in a comfortable key of their choice. LTAS was calculated for the first two phrases, the remainder of the song, and the entire song. The overall LTAS averages of each sample as well as the physiological and singing ranges were compared with self-reported singing-voice classification. Voice classification and overall LTAS average were moderately correlated, but the strength of the correlation varied with each sample. The strongest correlation was with the entire song. Voice classification and singing range were strongly correlated. LTAS remains a promising tool to aid in singing-voice classification. However, how to best use LTAS in classification remains unclear because of the influence of sample length and phonetic and pitch content on LTAS.