Ομιλία

Use of the Harmonic Phase in several Speech Processing Applications
27.06.2011
Ομιλητής: Inma Hernaez
Ημερομηνία: Δευτέρα 27 Ιουνίου, 2011 Ώρα: 11:45 - 13:30
Τοποθεσία: Aίθουσα Συναντήσεων "Στέλιος Ορφανουδάκης", ΙΤΕ, Ηράκλειο, Κρήτη
Host: Γιάννης Στυλιανού

Abstract:

In this talk the development of a new parameterization technique for the harmonic phase of the speech signal will be described. The harmonic phase is derived from the classical Harmonic plus Stocastic model, and simple manipulations over the original phase are performed, leading to a nice and meaningful representation named Relative Phase Shift. These values have been used in several speech processing applications, such as polarity detection, Speech and Speaker recognition and detection of synthetic speech signals. Some experiments and results will be described in the talk.

Additionally, the history and research activities at the Aholab Signal Processing Laboratory of the University of the Basque Country will be presented.

Bio:

Inmaculada Hernaez Rioja holds an MS degree in Electrical Engineering from the Universidad Politecnica de Catalunya (UPC) and a Ph.D. degree from the Universidad del Pais Vasco (UPV/EHU) (1995), also in Electrical Engineering. Currently she is full professor of the University of the Basque Country, giving courses related to Signal Theory and Communications in the Telecommunications Engineering master and degree. She is a founding member of the Aholab Signal Processing Laboratory and presently leads the group activities. She is managing several national projects and has established strong collaboration with private firms and public institutions to provide knowledge and technology in speech applications as well as providing speech resources for the technological development of the Basque language. Her present interests in research include the use of the phase in different speech processing applications and the development of high quality speech vocoders.

Όροι Χρήσης | Πολιτική Προστασίας Δεδομένων Προσωπικού Χαρακτήρα