Abstract | ||
---|---|---|
In order to capture sequential information and to take advantage of extended training data conditions, we developed an algorithm for speaker detection that scores a test segment by comparing it directly to similar instances of that speech in the training data. This non-parametric technique, though at an early stage in its development, achieves error rates close to 1% on the NIST 2001 Extended Data task and performs extremely well in combination with a standard Gaussian Mixture Model system. We also present a new scoring method that significantly improves performance by capturing only positive evidence. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1109/ICASSP.2005.1415224 | 2005 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1-5: SPEECH PROCESSING |
Keywords | Field | DocType |
gaussian mixture model,hidden markov models,computer science,learning artificial intelligence,gmm,gaussian processes,sequential analysis,automatic speech recognition,training data,speaker recognition,error rate,loudspeakers,nist | Training set,Pattern recognition,Computer science,Speech recognition,Speaker recognition,NIST,Artificial intelligence,Gaussian process,Speaker detection,Loudspeaker,Hidden Markov model,Mixture model | Conference |
ISSN | Citations | PageRank |
1520-6149 | 3 | 0.41 |
References | Authors | |
6 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Daniel Gillick | 1 | 55 | 3.76 |
Stephen Stafford | 2 | 3 | 0.41 |
Barbara Peskin | 3 | 176 | 18.45 |