traineracousticmodel.java

It is the Speech recognition software. It is platform independent. To execute the source code,

/*
 * Copyright 1999-2002 Carnegie Mellon University.  
 * Portions Copyright 2002 Sun Microsystems, Inc.  
 * Portions Copyright 2002 Mitsubishi Electric Research Laboratories.
 * All Rights Reserved.  Use is subject to license terms.
 * 
 * See the file "license.terms" for information on usage and
 * redistribution of this file, and for a DISCLAIMER OF ALL 
 * WARRANTIES.
 *
 */

package edu.cmu.sphinx.linguist.acoustic.tiedstate.trainer;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.util.logging.Logger;

import edu.cmu.sphinx.linguist.acoustic.tiedstate.Loader;
import edu.cmu.sphinx.linguist.acoustic.tiedstate.Saver;
import edu.cmu.sphinx.linguist.acoustic.tiedstate.TiedStateAcousticModel;
import edu.cmu.sphinx.util.SphinxProperties;
import edu.cmu.sphinx.util.Timer;

/**
 * Represents the generic interface to the Acoustic 
 * Model for sphinx4
 */
public class TrainerAcousticModel extends TiedStateAcousticModel {


    /**
     * Prefix for acoustic model SphinxProperties.
     */
    public final static String PROP_PREFIX
	= "edu.cmu.sphinx.linguist.acoustic.";


    /**
     * The directory where the acoustic model data can be found.
     */
    public final static String PROP_LOCATION_SAVE = 
	PROP_PREFIX + "location.save";


    /**
     * The default value of PROP_LOCATION_SAVE.
     */
    public final static String PROP_LOCATION_SAVE_DEFAULT = ".";


    /**
     * The save format for the acoustic model data. Current supported
     * formats are:
     *
     *  sphinx3.ascii
     *  sphinx3.binary
     *  sphinx4.ascii
     *  sphinx4.binary
     */
    public final static String PROP_FORMAT_SAVE = PROP_PREFIX + "format.save";


    /**
     * The default value of PROP_FORMAT_SAVE.
     */
    public final static String PROP_FORMAT_SAVE_DEFAULT = "sphinx3.binary";


    /**
     * The file containing the phone list.
     */
    public final static String PROP_PHONE_LIST = "phone_list";


    /**
     * The default value of PROP_PHONE_LIST.
     */
    public final static String PROP_PHONE_LIST_DEFAULT = "phonelist";


    /**
     * Flag indicating all models should be operated on.
     */
    public final static int ALL_MODELS = -1;

    /**
     * The logger for this class
     */
    private static Logger logger = 
	    Logger.getLogger(PROP_PREFIX + "TrainerAcousticModel");

    /**
     * The pool manager
     */
    private HMMPoolManager hmmPoolManager;
    private Loader loader;


    /**
     * Initializes the acoustic model
     *
     * @throws IOException if the model could not be created
     */
    public void initialize(String name, String context) throws IOException {
	loader = new ModelInitializerLoader(name, props);
	hmmPoolManager = new HMMPoolManager(loader, props);
	this.name = name;
	this.context = context;
        this.props = SphinxProperties.getSphinxProperties(context);
        this.loadTimer = Timer.getTimer(context, TIMER_LOAD);
	logInfo();
    }


    /**
     * Saves the acoustic model with a given name and format
     *
     * @param name the name of the acoustic model
     *
     * @throws IOException if the model could not be loaded
     * @throws FileNotFoundException if the model does not exist
     */
    public void save(String name) throws IOException, FileNotFoundException {
	Saver saver;

	String formatProp = PROP_FORMAT_SAVE;
	if (name != null) {
	    formatProp = PROP_PREFIX + name + ".format.save";
	}
	String format = props.getString(formatProp, PROP_FORMAT_SAVE_DEFAULT);

        if (format.equals("sphinx3.ascii")) {
	    logger.info("Sphinx-3 ASCII format");
            saver = new Sphinx3Saver(name, props, false, loader);
        } else if (format.equals("sphinx3.binary")) {
	    logger.info("Sphinx-3 binary format");
            saver = new Sphinx3Saver(name, props, true, loader);
        }  else if (format.equals("sphinx4.ascii")) {
	    logger.info("Sphinx-4 ASCII format");
            saver = new Sphinx4Saver(name, props, false, loader);
        } else if (format.equals("sphinx4.binary")) {
	    logger.info("Sphinx-4 binary format");
            saver = new Sphinx4Saver(name, props, true, loader);
        }  else { // add new saving code here.
            saver = null;
            logger.severe("Unsupported acoustic model format " + format);
	}
    }

    /**
     * Loads the acoustic models. This has to be explicitly requested
     * in this class.
     *
     * @throws IOException if the model could not be loaded
     * @throws FileNotFoundException if the model does not exist
     */
    public void load() throws IOException, FileNotFoundException {
        loadTimer.start();
        super.load();
        loadTimer.stop();
        logInfo();
	hmmPoolManager = new HMMPoolManager(loader, props);
    }

    /**
     * Resets the buffers.
     */
    public void resetBuffers() {
	// Resets the buffers and associated variables.
	hmmPoolManager.resetBuffers();
    }


    /**
     * Accumulate the current TrainerScore into the buffers.
     *
     * @param index the current index into the TrainerScore vector
     * @param trainerScore the TrainerScore in the current frame
     * @param nextTrainerScore the TrainerScore in the next frame
     */
    public void accumulate(int index, TrainerScore[] trainerScore, 
			   TrainerScore[] nextTrainerScore) {
	hmmPoolManager.accumulate(index, trainerScore, nextTrainerScore);
    }

    /**
     * Accumulate the current TrainerScore into the buffers.
     *
     * @param index the current index into the TrainerScore vector
     * @param trainerScore the TrainerScore
     */
    public void accumulate(int index, TrainerScore[] trainerScore) {
	hmmPoolManager.accumulate(index, trainerScore);
    }

    /**
     * Update the log likelihood. This should be called at the end of
     * each utterance.
     */
    public void updateLogLikelihood() {
	hmmPoolManager.updateLogLikelihood();
    }

    /**
     * Normalize the buffers and update the models.
     *
     * @return the log likelihood for the whole training set
     */
    public float normalize() {
	float logLikelihood = hmmPoolManager.normalize();
	hmmPoolManager.update();
	return logLikelihood;
    }


}