Title :
Frame-based acoustic feature integration for speech understanding
Author :
Barrault, Loic ; Servan, Christophe ; Matrouf, Driss ; Linarès, Georges ; de Mori, Renato
Author_Institution :
LIA, Avignon Univ., Avignon
fDate :
March 31 2008-April 4 2008
Abstract :
With the purpose of improving spoken language understanding (SLU) performance, a combination of different acoustic speech recognition (ASR) systems is proposed. State a posteriori probabilities obtained with systems using different acoustic feature sets are combined with log-linear interpolation. In order to perform a coherent combination of these probabilities, acoustic models must have the same topology (i.e. same set of states). For this purpose, a fast and efficient twin model training protocol is proposed. By a wise choice of acoustic feature sets and log-linear interpolation of their likelihood ratios, a substantial concept error rate (CER) reduction has been observed on the test part of the French MEDIA corpus.
Keywords :
interpolation; speech recognition; French MEDIA corpus; acoustic feature sets; acoustic speech recognition systems; automatic speech recognition; concept error rate reduction; frame-based acoustic feature integration; log-linear interpolation; speech understanding; spoken language understanding; state a posteriori probabilities; twin model training protocol; Acoustic testing; Automatic speech recognition; Decoding; Error analysis; Interpolation; Lattices; Natural languages; Protocols; Speech recognition; Topology; frame based combination; posterior probabilities combination; speech recognition; speech understanding;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4244-1483-3
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2008.4518780