Abstract | ||
---|---|---|
We describe a speaker-cluster normalization algorithm that we applied to both gender-normalization and speaker-normalization. To achieve parameter sharing the acoustic space is partitioned into classes. A maximum likelihood approach has been proposed under which the data between the distribution mean and its corresponding acoustic class is mostly speaker-independent, whereas the means of the acoustic classes are mostly speaker-dependent. When applied to gender-normalization the error rate reduction approaches that of a gender-dependent system but with half the number of parameters. For a speaker-normalized system, a 30% decrease in error rate was obtained in a batch recognition experiment in a context-dependent continuous-density HMM system. |
Year | DOI | Venue |
---|---|---|
1996 | 10.1109/ICASSP.1996.541102 | ICASSP |
Keywords | Field | DocType |
context-dependent continuous-density hmm system,error rate reduction approach,acoustic class,continuous-density hidden markov model,error rate,gender-dependent system,distribution mean,batch recognition experiment,acoustic space,speaker-normalized system,corresponding acoustic class,gender normalization,training data,maximum likelihood,acoustic noise,speech recognition,hidden markov models,loudspeakers,context dependent,convergence,speech processing,maximum likelihood estimation | Convergence (routing),Noise,Speech processing,Normalization (statistics),Pattern recognition,Computer science,Word error rate,Artificial intelligence,Acoustic space,Loudspeaker,Hidden Markov model | Conference |
ISBN | Citations | PageRank |
0-7803-3192-3 | 13 | 1.53 |
References | Authors | |
9 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
A. Acero | 1 | 4390 | 478.73 |
Xuedong Huang | 2 | 1390 | 283.19 |