Bitte benutzen Sie diese Referenz, um auf diese Ressource zu verweisen:
doi:10.22028/D291-25193
Titel: | A new model-discriminant training algorithm for hybrid NN-HMM systems |
VerfasserIn: | Reichl, W. Caspary, P. Ruske, G. |
Sprache: | Englisch |
Erscheinungsjahr: | 1996 |
Quelle: | Saarbrücken, 1996 |
Kontrollierte Schlagwörter: | Künstliche Intelligenz |
DDC-Sachgruppe: | 004 Informatik |
Dokumenttyp: | Forschungsbericht (Report zu Forschungsprojekten) |
Abstract: | This paper describes a hybrid system for continuous speech recognition consisting of a neural network (NN) and a hidden Markov model (HMM). The system is based on a multilayer perceptron, which approximates the a-posteriori probability of a sequence of states, derived from semi-continuous hidden Markov models. The classification is based on a total score for each hybrid model, attained from a Viterbi search on the state probabilities. Due to the unintended discrimination between the states in each model, a new training algorithm for the hybrid neural networks is presented. The utilized error function approximates the misclassification rate of the hybrid system. The discriminance between the correct and the incorrect models is optimized during the training by the "Generalized Probabilistic Descent Algorithm';, resulting in a minimum classification error. No explicit target values for the neural net output nodes are used, as in the usual backpropagation algorithm with a quadratic error function. In basic experiments up to 56% recognition rate were achieved on a vowel classification task and up to 69 % on a consonant cluster classification task. |
Link zu diesem Datensatz: | urn:nbn:de:bsz:291-scidok-41900 hdl:20.500.11880/25249 http://dx.doi.org/10.22028/D291-25193 |
Schriftenreihe: | Vm-Report / Verbmobil, Verbundvorhaben, [Deutsches Forschungszentrum für Künstliche Intelligenz] |
Band: | 108 |
Datum des Eintrags: | 6-Sep-2011 |
Fakultät: | SE - Sonstige Einrichtungen |
Fachrichtung: | SE - DFKI Deutsches Forschungszentrum für Künstliche Intelligenz |
Sammlung: | SciDok - Der Wissenschaftsserver der Universität des Saarlandes |
Dateien zu diesem Datensatz:
Datei | Beschreibung | Größe | Format | |
---|---|---|---|---|
report_108_96.pdf | 294,07 kB | Adobe PDF | Öffnen/Anzeigen |
Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt.