Abstract | ||
---|---|---|
Word level information obtained from the output of a speech recognizer has been used in the past to extract confidence features for the hypothesized words. This work describes a post-recognition process which treats these word-level features as independent knowledge sources and combines them in one log linear model for the posterior probability of a word sequence. This model is used for rescoring the hypotheses. The parameters of the model are optimized using a discriminative model combination approach, where a simplex optimization method, known as amoeba search, is used to minimize the non-smooth function of empirical error rate on training data. The method is evaluated on the SWITCHBOARD database. After training 20 new parameters, we obtain a significant word error rate reduction over the baseline system. A correlation measure between features and word accuracy is defined to help analyze and explain the results. |
Year | DOI | Venue |
---|---|---|
2000 | 10.1109/ICASSP.2000.862109 | 2000 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, PROCEEDINGS, VOLS I-VI |
Keywords | Field | DocType |
discriminative model,training data,error rate,feature extraction,natural languages,log linear model,data mining,speech processing,speech recognition,posterior probability,word error rate | Factored language model,Simplex algorithm,Pattern recognition,Computer science,Word error rate,Feature extraction,Speech recognition,Posterior probability,Correlation,Artificial intelligence,Log-linear model,Discriminative model | Conference |
ISSN | Citations | PageRank |
1520-6149 | 7 | 0.83 |
References | Authors | |
2 | 1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dimitra Vergyri | 1 | 373 | 36.97 |