Title :
Context-Dependent Duration Modeling
Author_Institution :
Temic Speech Dialog Syst., Ulm, Germany
fDate :
March 18-23, 2005
Keywords :
Viterbi decoding; hidden Markov models; speech recognition; 1-best Viterbi decoder; conditional duration probabilities; connected digit recognition; context-dependent duration model; duration-dependent density functions; first-order HMM-based speech recognizer; hidden Markov models; relative word error reduction; Context modeling; Decoding; Density functional theory; Distribution functions; Hidden Markov models; Probability distribution; Solid modeling; Speech recognition; Topology; Viterbi algorithm;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on
Print_ISBN :
0-7803-8874-7
DOI :
10.1109/ICASSP.2005.1415140
