A fuzzy classifier to deal with similarity between labels on automatic prosodic labeling |
| |
| Affiliation: | 1. Departamento de Farmacia, Facultad de Química, Universidad Nacional Autónoma de México, Av. Universidad 3000, Copilco Universidad, Coyoacán, 04510 Ciudad de México, Distrito Federal, México D.F., Mexico;2. Intituto de Quimica, Universidad Nacional Autónoma de México, Av. Universidad 3000, Copilco Universidad, Coyoacán, 04510 Ciudad de México, Distrito Federal, México D.F., Mexico;3. Departamento de Sistemas Biológicos y de Producción Agrícola y Animal, Universidad Autónoma Metropolitana-Xochimilco, Calzada del Hueso 1100, Col. Villa Quietud, 04960 Ciudad de México, México D.F. C.P. 04960, Mexico;4. Departamento de Patología, Hospital General de México, Dr. Balmis 148, México D.F., Mexico;5. Faculty of Chemistry, Department of Pharmacy (Departamento de Farmacia), National Autonomous University of Mexico in Mexico City (Universidad Nacional Autónoma de México), Mexico;1. Université de Moncton, Shippagan Campus, New Brunswick E8S 1P6, Canada;2. Department of Computer Engineering, King Saud University, Riyadh, Saudi Arabia;3. Department of French, University of New Brunswick, Canada;4. Department of Computer and Electrical Engineering, UQTR, Canada;5. Faculty of Electronics and Computer Science, USTHB University, Algiers, Algeria;1. Suez Canal University, Cairo, Egypt;2. University of Quebec in Montreal (UQAM), 201 President Kennedy, Montreal, QC, Canada, H2X 3Y7;1. LISI Laboratory of Computer Science for Industrial Systems, Higher Institute of Documentation (ISD), Manouba University 2010, Tunisia;2. RIADI Research Laboratory, ENSI, Manouba University 2010, Tunisia;3. Emirates College of Technology, P.O. Box: 41009, Abu Dhabi, United Arab Emirates;4. Informatics Research Institute of Toulouse (IRIT), 02 Rue de Charles Camichel, 31071 Toulouse Cedex 7, France;5. Higher Institute of Informatics (ISI), Tunis El Manar University, 2080 Ariana, Tunisia;1. Center for Spoken Language Understanding, Oregon Health & Science University (OHSU), Portland, OR, United States;2. Oregon Center for Aging & Technology, Oregon Health & Science University (OHSU), Portland, OR, United States |
| |
| Abstract: | This paper presents an original approach to automatic prosodic labeling. Fuzzy logic techniques are used for representing situations of high uncertainty with respect to the category to be assigned to a given prosodic unit. The Fuzzy Integer technique is used to combine the output of different base classifiers. The resulting fuzzy classifier benefits from the different capabilities of the base classifiers for identifying different types of prosodic events. At the same time, the fuzzy classifier identifies the events that are potentially more difficult to be labeled. The classifier has been applied to the identification of ToBI pitch accents. The state of the art on pitch accent multiclass classification reports around 70% accuracy rate. In this paper we describe a fuzzy classifier which assigns more than one label in confusing situations. We show that the pairs of labels that appear in these uncertain situations are consistent with the most confused pairs of labels reported in manual prosodic labeling experiments. Our fuzzy classifier obtains a soft classification rate of 81.8%, which supports the potential of the proposed system for computer assisted prosodic labeling. |
| |
| Keywords: | Automatic prosodic labeling Fuzzy classification ToBI |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
