sphinx3loader.java

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

/*
 * Copyright 1999-2004 Carnegie Mellon University.  
 * Portions Copyright 2004 Sun Microsystems, Inc.  
 * Portions Copyright 2004 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;

import java.io.BufferedInputStream;
import java.io.DataInputStream;
import java.io.File;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InputStream;

import java.net.URL;

import java.util.LinkedHashMap;
import java.util.Map;
import java.util.Properties;
import java.util.logging.Level;
import java.util.logging.Logger;
import java.util.zip.ZipException;

import edu.cmu.sphinx.linguist.acoustic.Context;
import edu.cmu.sphinx.linguist.acoustic.HMM;
import edu.cmu.sphinx.linguist.acoustic.HMMPosition;
import edu.cmu.sphinx.linguist.acoustic.LeftRightContext;
import edu.cmu.sphinx.linguist.acoustic.Unit;
import edu.cmu.sphinx.linguist.acoustic.UnitManager;

import edu.cmu.sphinx.linguist.acoustic.tiedstate.*;

import edu.cmu.sphinx.util.ExtendedStreamTokenizer;
import edu.cmu.sphinx.util.LogMath;
import edu.cmu.sphinx.util.SphinxProperties;
import edu.cmu.sphinx.util.StreamFactory;
import edu.cmu.sphinx.util.Utilities;
import edu.cmu.sphinx.util.props.PropertyException;
import edu.cmu.sphinx.util.props.PropertySheet;
import edu.cmu.sphinx.util.props.PropertyType;
import edu.cmu.sphinx.util.props.Registry;

/**
 * Loads a tied-state acoustic model generated by the Sphinx-3 trainer.
 * <p>
 * The acoustic model should be packaged in a JAR file.
 * The dictionary and language model files are not required
 * to be in the package. You can specify their locations separately.
 * A text file called "model.props" and the data files that make up the 
 * acoustic model are required.
 * The model.props file is a file of key-value pairs, loadable as a Java
 * Properties file. It should minimally contain the following properties:
 * <ul>
 * <li><b>dataLocation</b> - this specifies the directory where the 
 *     actual model data files are, <i>relative to the model implementation 
 *     class</i></li>
 * <li><b>modelDefinition</b> - this specifies the location where the 
 *     model definition file is, <i>relative to the model implementation 
 *     class</i></li> 
 * </ul>
 * The actual model data files are named "means", "variances", 
 * "transition_matrices", "mixture_weights" for binary versions, 
 * or prepended with ".ascii" for the ASCII versions.
 * </p>
 * <p>
 * As an example, lets look at the Wall Street Journal acoustic model JAR 
 * file, which is located at the <code>sphinx4/lib</code> directory. 
 * If you run 
 * <code>"jar tvf lib/WSJ_8gau_13dCep_16k_40mel_130Hz_6800Hz.jar"</code>,
 * you will find that its internal structure looks roughly like:
 * <pre>
 * WSJ_8gau_13dCep_16k_40mel_130Hz_6800Hz.jar
 *   |
 *   +- edu
 *       |
 *       +- cmu
 *           |
 *           +- sphinx
 *               |
 *               +- model
 *                   |
 *                   + acoustic
 *                      |
 *                      +- model.props
 *                      |
 *                      +- WSJ_8gau_13dCep_16k_40mel_130Hz_6800Hz.class
 *                      |
 *                      +- WSJLoader.class
 *                      |
 *                      +- cd_continuous_8gau
 *                      |   |
 *                      |   +- means
 *                      |   +- variances
 *                      |   +- mixture_weights
 *                      |   +- transition_matrices
 *                      |
 *                      +- dict
 *                      |   |
 *                      |   +- alpha.dict
 *                      |   +- cmudict.0.6d
 *                      |   +- digits.dict
 *                      |   +- fillerdict
 *                      |
 *                      +- etc
 *                          |
 *                          +- WSJ_clean_13dCep_16k_40mel_130Hz_6800Hz.4000.mdef
 *                          +- WSJ_clean_13dCep_16k_40mel_130Hz_6800Hz.ci.mdef
 *                          +- variables.def
 *
 * </pre>
 * <p>
 * The model.props file looks like (note how the 'dataLocation' and
 * 'modelDefinition' properties are defined relative to the 
 * WSJ_clean_13dCep_16k_40mel_130Hz_6800Hz.class):
 * </p>
 * <pre>
description = Wall Street Journal acoustic models

isBinary = true
featureType = cepstra_delta_doubledelta
vectorLength = 39
sparseForm = false

numberFftPoints = 512
filters = 40
gaussians = 8
maxFreq = 6800
minFreq. = 130
sampleRate = 16000

dataLocation = cd_continuous_8gau
modelDefinition = etc/WSJ_clean_13dCep_16k_40mel_130Hz_6800Hz.4000.mdef
 * </pre>
 *
 * <p>
 * Note that although most of the properties of this class are already
 * defined in the model.props file, it is still possible (but not recommended)
 * to override those values by specifying them in the configuration file.
 * </p>
 */
public class Sphinx3Loader implements Loader {
    /**
     * The log math component for the system.
     */
    public final static String PROP_LOG_MATH = "logMath";

    /**
     * The unit manager
     */
    public final static String PROP_UNIT_MANAGER  = "unitManager";

    /**
     * Specifies whether the model to be loaded is in ASCII or binary format
     */
    public final static String PROP_IS_BINARY = "isBinary";

    /**
     * The default value of PROP_IS_BINARY
     */
    public final static boolean PROP_IS_BINARY_DEFAULT = true;

    /**
     * The name of the model definition file (contains the HMM data)
     */
    public final static String PROP_MODEL = "modelDefinition";

    /**
     * The default value of PROP_MODEL_DEFAULT.
     */
    public final static String PROP_MODEL_DEFAULT = "model.mdef";

    /**
     * Subdirectory where the acoustic model can be found
     */
    public final static String PROP_DATA_LOCATION = "dataLocation";

    /**
     * The default value of PROP_DATA_LOCATION.
     */
    public final static String PROP_DATA_LOCATION_DEFAULT = "data";

    /**
     * The SphinxProperty for the name of the acoustic properties file.
     */
    public final static String PROP_PROPERTIES_FILE = "propertiesFile";

    /**
     * The default value of PROP_PROPERTIES_FILE.
     */
    public final static String PROP_PROPERTIES_FILE_DEFAULT = "model.props";

    /**
     * The SphinxProperty for the length of feature vectors.
     */
    public final static String PROP_VECTOR_LENGTH = "vectorLength";

    /**
     * The default value of PROP_VECTOR_LENGTH.
     */
    public final static int PROP_VECTOR_LENGTH_DEFAULT = 39;

    /**
     * The SphinxProperty specifying whether the transition matrices of the
     * acoustic model is in sparse form, i.e., omitting the zeros of the
     * non-transitioning states.
     */
    public final static String PROP_SPARSE_FORM = "sparseForm";

    /**
     * The default value of PROP_SPARSE_FORM.
     */
    public final static boolean PROP_SPARSE_FORM_DEFAULT = true;

    /**
     * The SphinxProperty specifying whether context-dependent units should be
     * used.
     */
    public final static String PROP_USE_CD_UNITS = "useCDUnits";

    /**
     * The default value of PROP_USE_CD_UNITS.
     */
    public final static boolean PROP_USE_CD_UNITS_DEFAULT = true;

    /**
     * Mixture component score floor.
     */
    public final static String PROP_MC_FLOOR = "MixtureComponentScoreFloor";

    /**
     * Mixture component score floor default value.
     */
    public final static float PROP_MC_FLOOR_DEFAULT = 0.0f;

    /**
     * Variance floor.
     */
    public final static String PROP_VARIANCE_FLOOR = "varianceFloor";

    /**
     * Variance floor default value.
     */
    public final static float PROP_VARIANCE_FLOOR_DEFAULT = 0.0001f;

    /**
     * Mixture weight floor.
     */
    public final static String PROP_MW_FLOOR = "mixtureWeightFloor";

    /**
     * Mixture weight floor default value.
     */
    public final static float PROP_MW_FLOOR_DEFAULT = 1e-7f;


    protected final static String NUM_SENONES = "num_senones";
    protected final static String NUM_GAUSSIANS_PER_STATE = "num_gaussians";
    protected final static String NUM_STREAMS = "num_streams";
    protected final static String FILLER = "filler";
    protected final static String SILENCE_CIPHONE = "SIL";
    protected final static int BYTE_ORDER_MAGIC = 0x11223344;
    /**
     * Supports this version of the acoustic model
     */
    public final static String MODEL_VERSION = "0.3";
    protected final static int CONTEXT_SIZE = 1;
    private Pool meansPool;
    private Pool variancePool;
    private Pool matrixPool;
    private Pool meanTransformationMatrixPool;
    private Pool meanTransformationVectorPool;
    private Pool varianceTransformationMatrixPool;
    private Pool varianceTransformationVectorPool;
    private Pool mixtureWeightsPool;
    private Pool senonePool;
    private Map contextIndependentUnits;
    private HMMManager hmmManager;
    private LogMath logMath;
    private UnitManager unitManager;
    private Properties properties;
    private boolean swap;
    protected final static String DENSITY_FILE_VERSION = "1.0";
    protected final static String MIXW_FILE_VERSION = "1.0";
    protected final static String TMAT_FILE_VERSION = "1.0";
    // --------------------------------------
    // Configuration variables
    // --------------------------------------
    private String name;
    private Logger logger;
    private boolean binary;
    private boolean sparseForm;
    private int vectorLength;
    private String location;
    private String model;
    private String dataDir;
    private String propsFile;
    private float distFloor;
    private float mixtureWeightFloor;
    private float varianceFloor;
    private boolean useCDUnits;


    /*
     * (non-Javadoc)
     * 
     * @see edu.cmu.sphinx.util.props.Configurable#register(java.lang.String,
     *      edu.cmu.sphinx.util.props.Registry)
     */
    public void register(String name, Registry registry)
            throws PropertyException {
        this.name = name;
        registry.register(PROP_LOG_MATH, PropertyType.COMPONENT);
        registry.register(PROP_UNIT_MANAGER, PropertyType.COMPONENT);
        registry.register(PROP_IS_BINARY, PropertyType.BOOLEAN);
	registry.register(PROP_SPARSE_FORM, PropertyType.BOOLEAN);
        registry.register(PROP_VECTOR_LENGTH, PropertyType.INT);
        registry.register(PROP_MODEL, PropertyType.STRING);
        registry.register(PROP_DATA_LOCATION, PropertyType.STRING);
        registry.register(PROP_PROPERTIES_FILE, PropertyType.STRING);
        registry.register(PROP_MC_FLOOR, PropertyType.FLOAT);
        registry.register(PROP_MW_FLOOR, PropertyType.FLOAT);
        registry.register(PROP_VARIANCE_FLOOR, PropertyType.FLOAT);
        registry.register(PROP_USE_CD_UNITS, PropertyType.BOOLEAN);
    }

    /*
     * (non-Javadoc)
     * 
     * @see edu.cmu.sphinx.util.props.Configurable#newProperties(edu.cmu.sphinx.util.props.PropertySheet)
     */
    public void newProperties(PropertySheet ps) throws PropertyException {
        logger = ps.getLogger();
        propsFile =
            ps.getString(PROP_PROPERTIES_FILE, PROP_PROPERTIES_FILE_DEFAULT);
        logMath = 
            (LogMath) ps.getComponent(PROP_LOG_MATH, LogMath.class);
        unitManager = 
            (UnitManager) ps.getComponent(PROP_UNIT_MANAGER,
                                          UnitManager.class);
        binary = 
            ps.getBoolean(PROP_IS_BINARY, getIsBinaryDefault());
	sparseForm = 
            ps.getBoolean(PROP_SPARSE_FORM, getSparseFormDefault());
        vectorLength = 
            ps.getInt(PROP_VECTOR_LENGTH, getVectorLengthDefault());
        model = 
            ps.getString(PROP_MODEL, getModelDefault());
        dataDir = 
            ps.getString(PROP_DATA_LOCATION, getDataLocationDefault()) + "/";
        distFloor = 
            ps.getFloat(PROP_MC_FLOOR, PROP_MC_FLOOR_DEFAULT);
        mixtureWeightFloor = 
            ps.getFloat(PROP_MW_FLOOR, PROP_MW_FLOOR_DEFAULT);
        varianceFloor = 
            ps.getFloat(PROP_VARIANCE_FLOOR, PROP_VARIANCE_FLOOR_DEFAULT);
        useCDUnits = 
            ps.getBoolean(PROP_USE_CD_UNITS, PROP_USE_CD_UNITS_DEFAULT);
    }

    private void loadProperties() {
        if (properties == null) {
            properties = new Properties();
            try {
                URL url = getClass().getResource(propsFile);
                // System.out.println(getClass() + " " + url);
                properties.load(url.openStream());
            } catch (IOException ioe) {
                ioe.printStackTrace();
            }
        }
    }


    /**
     * Returns whether the models are binary by default
     *
     * @return true if the models are binary by default
     */
    private boolean getIsBinaryDefault() {
        loadProperties();
        String binary = (String) properties.get(PROP_IS_BINARY);
        if (binary != null) {
            return (Boolean.valueOf(binary).equals(Boolean.TRUE));
        } else {
            return PROP_IS_BINARY_DEFAULT;
        }
    }

    /**
     * Returns whether the matrices are in sparse form by default.
     *
     * @return true if the matrices are in sparse form by default
     */
    private boolean getSparseFormDefault() {
        loadProperties();
        String sparse = (String) properties.get(PROP_SPARSE_FORM);
        if (sparse != null) {
            return (Boolean.valueOf(binary).equals(Boolean.TRUE));
        } else {
            return PROP_SPARSE_FORM_DEFAULT;
        }
    }

    /**
     * Returns the default vector length.
     */
    private int getVectorLengthDefault() {
        loadProperties();