ASR based on the analasys of the short-MelFrequencyCepstra time transform
Chapter in Scopus
- Additional Document Info
- View All
In this work, we propose to use as source of speech information the Short-MelfrequencyCepstra Time Transform (SMCTT), c¿(t). The SMCTT studies the time properties at quefrency ¿. Since the SM-CTT signal, c¿(t), comes from a nonlinear transformation of the speech signal, s(t), it makes the STMCTT a potential signal with new properties in time, frequency, quefrency, etc. The goal of this work is to present the performance of the SMCTT signal when the SMCTT is applied to an Automatic Speech Recognition (ASR) task. Our experiment results show that important information is given by this SMCTT waveform, c¿(t). © Springer-Verlag Berlin Heidelberg 2005.