Conference_programme: 18.3 - Voice accommodations in room acoustics and noise
Lecture: Singing voice quality assessment in professional singers using microphones and contact sensors
Author(s): Castellana Antonella, Carullo Alessio, Astolfi Arianna, Leocata Ilaria
Summary:
This paper investigates the singing voice quality through acoustic parameters obtained from three sensors: a sound level meter (SLM), placed on-axis at 30 cm of distance from the mouth, and two contact microphones that sense the vibration at the base of the neck, namely a piezoelectric (PM) and an electret condenser (ECM) microphones.\nFourteen professional singers (11 females and 3 males) participated to the experiment: in an anechoic chamber they were asked to sing one octave and a fifth arpeggio with two tempi and two articulations using the vowel /a/ and two song samples in a comfortable tonality, while standing in front of the SLM and wearing the PM and the ECM.\nSeveral singing voice quality parameters have been estimated: some related to pitch inaccuracy (the distance in cents in equally tempered tuning between the reference note and the sung note) and to the singer’s formant (Singing Power Ratio) in vowels, others related to spectral patterns (Long Term Average Spectrum and its relative features) in the song samples. \nThe pitch inaccuracy estimation was comparable for the three devices, as an average difference of 1.9 Hz between the estimated frequencies was found. As expected, spectral parameters differed a lot between the contact sensors and the SLM, since signals from contact sensors are not affected by the vocal tract filtering. However, for the professional tenors the singer’s formant was clearly noticeable in the spectra of the SLM but also of the PM and ECM. Such an effect was also detectable from contact sensors because it is a boost in middle-high frequencies that a trained voice obtains from adequately preparing his vocal apparatus to sing. Future research will investigate comparisons between spectral patterns in trained and untrained singers using the same contact sensors and will explore contact sensors as suitable devices for voice training monitoring.
Corresponding author
Name: Ms Antonella Castellana
e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Country: Italy
