Vol 3 (2020)

XXVII Pacific Voice Conference

Kraków, Poland (EU Edition)
October 24-26, 2019

Conference Proceedings

PUBLISHED: 2020-03-26

F. Mecorio, F. Santangelo, A.G. Gucciardo

Abstract: Pansori is a traditional Korean dramatic art which probably first appeared in the mid-eighteenth century in the southern region of Korea. “Han”, considered by many a word that cannot be translated into English, is the expression of a national sentiment of sorrow that plays a major role in pansori performances. However, almost no scholars have paid attention to the description of han in terms of acoustics and physiology, and the training of pansori itself still mostly relies on mimicking the sound of the teacher. In this pilot study, we would like to present our observations made of a small cluster of pansori performers expressing han. The preliminary findings of this pilot study may be useful as a reference for further research, both physiological and vocological.

Keywords: pansori, han. Estill, acoustics

Ž. Nedeljković, M. Milošević, Ž. Đurović

Abstract: Attempting to improve the performance of an automatic speech emotion recognition systems has been a daily endeavour. Diverse classification methods have been used as a part of this undertaking. It appears that HMM, a structure that was intensely exploited in previous research, is being rarely explored in new studies. We believe that it is precisely the particular structure of HMM that can introduce an additional quality into modern automatic emotion recognition systems, if not into a single classifier setup, then as a part of the multiclassifier structure. The aim of this paper is to improve discrete HMM as a first step in the process of successfully introducing it into modern automatic emotion recognition systems. The main contribution relates to the use of a new method of vector quantisation based on the QQ-plot. The proposed algorithm was tested, in combination with a wide set of features, on various emotional speech databases.

Keywords: Emotion recognition, HMM, QQ-plot

G. W. Johnson, DMA, H. S. Hunt, BA

Abstract: The initial 2004 research indicated that adult singers with a history of childhood sexual abuse (CSA) can present symptoms associated with posttrauma that are directly counterproductive to singing. Furthermore, symptoms of both posttrauma and therapeutic treatments for posttrauma can be impacted by the act of singing [1]. It was determined that optimal teaching/treatment strategies can occur best when the voice team (i.e. the singer and his/her teachers and health care providers) collaborate through an interdisciplinary exchange of information. Numerous aspects of the relationship of singing and CSA were identified as needing further study. Apart from ongoing one-on-one anecdotal communications from singers with a history of CSA, the research remained static for fourteen years. In 2018, further study was undertaken that expanded the existing knowledge base by incorporating information specific to childhood physical abuse (CPA) and violent sexual assault; and by creating a healing paradigm for singers [2].

Keywords: singing and sexual abuse, singing and physical abuse, singing and PTSD, healing paradigm

R. Britton

Abstract: The Alexander Technique (AT) is based on the idea that as vertebrates we are whole systems where all muscles are coordinated with each other, and that all muscles play a role in the movement of an animal. This includes the muscles involved with phonation. The principles of AT are summarized here with a singer in mind.

Keywords: Posture, Voice, Efficiency, Coordination, Mobility

M. Just, A. Racino, M.H. Tyc, M. Morawska-Kochman

Abstract: We describe an innovative high speed (HS) and high definition (HD) color system to visualize glottic activities. The system, called Advanced Larynx Imager System (ALIS), is designed by DiagNova Technologies, Wroclaw, Poland. ALIS features a laser light source and allows the simultaneous capture of visual and acoustic signals. Two capture regimes are available; HS (up to 4000 fps) or HD (up to 1024×1024 pixels) at the standard rate of 25 fps. Such an arrangement provides a quick and precise analysis of vocal folds kinematics, including kymographic data analysis.

Keywords: high-speed digital imaging, high definition, laser light, kymographs, DiagNova