Multi-scale piecewise quasi-stationary analysis of speech signals for robust speech recognition.

Vivek Tyagi - PhD Student - Multimedia Communications
Multimedia Communications

Date: -
Location: Eurecom

A fixed scale (typically 25ms) short time spectral analysis of speech signals, which are inherently multi-scale in nature (typically vowels last for 40-80ms while stops last for 10-20ms), is clearly sub-optimal for time-frequency resolution. In this work,