A Database of Laryngeal High-Speed Videos with Simultaneous High-Quality Audio Recordings of Pathological and Non-Pathological Voices

A Database of Laryngeal High-Speed Videos with Simultaneous High-Quality Audio Recordings of Pathological and Non-Pathological Voices P. Aichinger, I. Roesner, M. Leonhard, D.M. Denk-Linnert, W. Bigenzahn, B. Schneider-Stickler Division of Phoniatrics-Logopedics, Department of Otorhinolaryngology, Medical University of Vienna, Austria Waehringer Guertel 18-20, 1090 Vienna, Austria E-mail: philipp.aichinger@meduniwien.ac.at Abstract Auditory voice quality judgements are used intensively for the clinical assessment of pathological voice. Voice quality concepts are fuzzily defined and poorly standardized however, which hinders scientific and clinical communication. The described database documents a wide variety of pathologies and is used to investigate auditory voice quality concepts with regard to phonation mechanisms. The database contains 375 laryngeal high-speed videos and simultaneous high-quality audio recordings of sustained phonations of 80 pathological and 40 non-pathological subjects. Interval wise annotations regarding video and audio quality, as well as voice quality ratings are provided. Video quality is annotated for the visibility of anatomical structures and artefacts such as blurring or reduced contrast. Voice quality annotations include ratings on the presence of dysphonia and diplophonia. The purpose of the database is to aid the formulation of observationally well-founded models of phonation and the development of model-based automatic detectors for distinct types of phonation, especially for clinically relevant nonmodal voice phenomena. Another application is the training of audio-based fundamental frequency extractors on video-based reference fundamental frequencies. Keywords: pathological voice, dysphonia, diplophonia, laryngeal high-speed videos 1. Introduction Auditory voice quality judgements are used intensively for the clinical assessment of pathologic voice. These judgements aid the indication, selection, evaluation and optimization of clinical treatment techniques. Voice quality types that are typically described include dysphonia, diplophonia and euphonia. The term dysphonia includes all types of auditory divergences from normal voice quality. Diplophonia is a concept subordinate to dysphonia, i.e. the simultaneous presence of two pitches in the voice sound. The term euphonia is the antonym to dysphonia. Voice quality concepts are fuzzily defined and poorly standardized however, which hinders scientific and clinical communication and updates of terminology are needed (Gerrat & Kreiman, 2001). These problems are tackled by using a data corpus including voice quality annotations, which aids the formulation of observationally well-founded models of voice production and perception, and consequently updates in terminology. We propose to observe voice production by means of laryngeal high-speed imaging with simultaneous high-quality audio recordings that enable acoustic and perceptual analyses. In contrast to what has often been practiced in the field of disordered voice research (Olszewski, Shen & Jiang, 2011), video and audio quality criteria that allow for reproducible and non-destructive pre-selection of voice samples are used. State-of-the-art audio equipment and acquisition strategies from the scientific field of audio engineering are deployed into the field of clinical voice research. Professional condenser microphones with symmetrical wiring and a high-quality recorder are used with a sampling rate of 48 kHz and a quantisation resolution of 24 bits. Interval wise voice quality annotations accompany subject global ratings. 2. Data collection One hundred and twenty subjects of three clinical voice quality groups representing subject global ratings are …

Content truncated to 500 words