Title :
A speaker-stress resistant HMM isolated word recognizer
Author :
Paul, Douglas B.
Author_Institution :
Massachusetts Institute of Technology, Lexington, Massachusetts
Abstract :
Most current speech recognition systems are sensitive to variations in speaker style, the following is the result of an effort to make a Hidden Markov Model (HMM) Isolated Word Recognizer (IWR) tolerant to such speech changes caused by speaker stress. More than an order-of-magnitude reduction of the error rate was achieved for a 105 word simulated stress database and a 0% error rate was achieved for the TI 20 isolated word database.
Keywords :
Databases; Error analysis; Hidden Markov models; Immune system; Isolation technology; Laboratories; Speech recognition; Stress; Testing; Training data;
Conference_Titel :
Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '87.
DOI :
10.1109/ICASSP.1987.1169551
