Dear all, in the speech parameter generation algorithms for HMM-Based Speech Synthesis (keiichi Tokuda , takayoshi Yoshimura , takashi masuko , takao Kobayashi,tadashi kitamura) there is a figure (#2) named Spectra obtained from 1-mixture HMMs and 8-mixture HMMs what did this figure try to improve ? i know that it try to show the spectra generated from 1mix become more clearer in the formant structure than the one generated from 8 mix i'am wondering about the y-axis (log magnitude (db) and the x-axis is the frequency (khz) log magnitude (db) of what. what is the (db) ? what numbers of frames should i use to draw this spectra ?coz i found in the sptk tutorial that the starting (s) and ending frame (e) is assigned the same value 80. does this mean that this spectra was drawn for only one frame. i need realy to understand the meaning of this figure, dimensions used , the curve that has been drawn why sometimes it is shifted down in some position in 2 mix than the one of 1 mix Thanks Invite your mail contacts to join your friends list with Windows Live Spaces. It's easy! Try it! |