Abstract | ||
---|---|---|
Previous studies have shown that the recognition accuracyoften severely degrades at higher speech rates, whichcan basically be traced back to two main dimensions:acoustic and phonemic. Reasons for this effect can befound in the phonemic field (e.g. elisions) as well as on theacoustic level: with increasing rates of speech the spectralcharacteristics are changing. A main obstacle in thiscontext is the training data, consisting of only a smallfraction of samples, which can be labeled as... |
Year | Venue | Keywords |
---|---|---|
1999 | EUROSPEECH | data consistency,hidden markov model |
Field | DocType | Citations |
Maximum-entropy Markov model,Forward algorithm,Pattern recognition,Markov model,Computer science,Speech recognition,Artificial intelligence,Variable-order Markov model,Markov blanket,Cluster analysis,Hidden Markov model,Hidden semi-Markov model | Conference | 1 |
PageRank | References | Authors |
0.37 | 4 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Robert Faltlhauser | 1 | 26 | 3.62 |
Thilo Pfau | 2 | 113 | 15.74 |
Günther Ruske | 3 | 154 | 36.13 |