Title :
Robust and efficient confidence measure for isolated command recognition
Author :
Hernández-Ábrego, Gustavo ; Menéndez-Pidal, Xavier ; Olorenshaw, Lex
Author_Institution :
Spoken Language Technol., Sony NSCA, San Jose, CA, USA
Abstract :
A new confidence measure for isolated command recognition is presented. It is versatile and efficient in two ways. First, it is based exclusively on the speech recognizer´s output. In addition, it is robust to changes in the vocabulary, acoustic model and parameter settings. Its calculation is very simple and it is based on the computation of a pseudo-filler score from an N-best list. Performance is tested in two different command recognition applications. Finally, it is efficient to separate correct results both from incorrect ones and from false alarms caused by out-of-vocabulary elements and noise.
Keywords :
acoustic noise; acoustic signal processing; natural language interfaces; speech recognition; N-best list; acoustic model; confidence measure; isolated command recognition; out-of-vocabulary elements; parameter settings; pseudo-filler score; speech recognizer; vocabulary; Acoustic beams; Acoustic testing; Character recognition; Collision mitigation; Isolation technology; Natural languages; Robustness; Speech processing; Speech recognition; Vocabulary;
Conference_Titel :
Automatic Speech Recognition and Understanding, 2001. ASRU '01. IEEE Workshop on
Print_ISBN :
0-7803-7343-X
DOI :
10.1109/ASRU.2001.1034681
