edu.cmu.sphinx.result
Class PivotSausageMaker
java.lang.Object
edu.cmu.sphinx.result.AbstractSausageMaker
edu.cmu.sphinx.result.PivotSausageMaker
- All Implemented Interfaces:
- ConfidenceScorer, Configurable
public class PivotSausageMaker
- extends AbstractSausageMaker
This is an implementation of an alternative sausage making algorithm as described in the following paper. Runs in
less time than the SausageMaker, but will also build different sausages. A general algorithm for word graph matrix
decomposition, Hakkani-Tur, D. Riccardi, G., AT&T Labs.-Res., USA;, This paper appears in: 2003 IEEE International
Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03).
- Author:
- pgorniak
Method Summary |
Sausage |
makeSausage()
Turn the lattice contained in this sausage maker into a sausage object. |
ConfidenceResult |
score(Result result)
Computes confidences for a Result and returns a ConfidenceResult, a compact representation of all the hypothesis
contained in the result together with their per-word and per-path confidences. |
Methods inherited from class java.lang.Object |
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
PivotSausageMaker
public PivotSausageMaker()
makeSausage
public Sausage makeSausage()
- Turn the lattice contained in this sausage maker into a sausage object.
- Returns:
- the sausage producing by collapsing the lattice.
score
public ConfidenceResult score(Result result)
- Description copied from interface:
ConfidenceScorer
- Computes confidences for a Result and returns a ConfidenceResult, a compact representation of all the hypothesis
contained in the result together with their per-word and per-path confidences.
- Parameters:
result
- the result to compute confidences for
- Returns:
- a confidence result
- See Also:
ConfidenceScorer.score(edu.cmu.sphinx.result.Result)