RT Conference Proceedings
T1 Morphological processing of a dynamic compressive gammachirp filterbank for automatic speech recognition
A1 Cadore, Joyner
A1 Peláez Moreno, Carmen
A1 Gallardo Antolín, Ascensión
AB The Dynamic Compressive Gammachirp is presented for producing auditory-inspired feature extraction in Automatic Speech Recognition. The proposed acoustic features combine spectral subtraction and two-dimensional non-linear filtering technique most usually employed for image processing: morphological filtering. These features have been proven to be more robust to noisy speech than those based on simpler auditory filterbanks like the classical mel-scaled triangular filterbank, the Gammatone filterbank and the passive Gammachirp in a noisy Isolet database.
PB Universidad Autonoma De Madrid. Escuela Politécnica Superior
SN 978-84-616-1535-2
YR 2012
FD 2012
LK https://hdl.handle.net/10016/21642
UL https://hdl.handle.net/10016/21642
LA eng
NO Actas de: VII Jornadas en Tecnología del Habla and III Iberian SLTECH Workshop (IberSPEECH 2012). Madrid, 21-23 noviembre 2012.
NO This work has been partially supported by the Spanish Ministry of Science and Innovation CICYT Projects No. TEC2008-06382/TEC and No. TEC2011-26807.
DS e-Archivo
RD 1 jul. 2024