1 /* -*- c-basic-offset: 4; indent-tabs-mode: nil -*- */ 2 /* ==================================================================== 3 * Copyright (c) 1999-2004 Carnegie Mellon University. All rights 4 * reserved. 5 * 6 * Redistribution and use in source and binary forms, with or without 7 * modification, are permitted provided that the following conditions 8 * are met: 9 * 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 13 * 2. Redistributions in binary form must reproduce the above copyright 14 * notice, this list of conditions and the following disclaimer in 15 * the documentation and/or other materials provided with the 16 * distribution. 17 * 18 * This work was supported in part by funding from the Defense Advanced 19 * Research Projects Agency and the National Science Foundation of the 20 * United States of America, and the CMU Sphinx Speech Consortium. 21 * 22 * THIS SOFTWARE IS PROVIDED BY CARNEGIE MELLON UNIVERSITY ``AS IS'' AND 23 * ANY EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, 24 * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 25 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL CARNEGIE MELLON UNIVERSITY 26 * NOR ITS EMPLOYEES BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, 27 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT 28 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 29 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 30 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 31 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE 32 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 33 * 34 * ==================================================================== 35 * 36 */ 37 /* 38 * ms_mgau.h -- Essentially a wrapper that wrap up gauden and 39 * senone. It supports multi-stream. 40 * 41 * 42 * ********************************************** 43 * CMU ARPA Speech Project 44 * 45 * Copyright (c) 1997 Carnegie Mellon University. 46 * ALL RIGHTS RESERVED. 47 * ********************************************** 48 * HISTORY 49 * $Log$ 50 * Revision 1.1 2006/04/05 20:27:30 dhdfu 51 * A Great Reorganzation of header files and executables 52 * 53 * Revision 1.3 2006/02/22 16:57:15 arthchan2003 54 * Fixed minor dox-doc issue 55 * 56 * Revision 1.2 2006/02/22 16:56:01 arthchan2003 57 * Merged from SPHINX3_5_2_RCI_IRII_BRANCH: Added ms_mgau.[ch] into the trunk. It is a wrapper of ms_gauden and ms_senone 58 * 59 * Revision 1.1.2.4 2005/09/25 18:55:19 arthchan2003 60 * Added a flag to turn on and off precomputation. 61 * 62 * Revision 1.1.2.3 2005/08/03 18:53:44 dhdfu 63 * Add memory deallocation functions. Also move all the initialization 64 * of ms_mgau_model_t into ms_mgau_init (duh!), which entails removing it 65 * from decode_anytopo and friends. 66 * 67 * Revision 1.1.2.2 2005/08/02 21:05:38 arthchan2003 68 * 1, Added dist and mgau_active as intermediate variable for computation. 2, Added ms_cont_mgau_frame_eval, which is a multi stream version of GMM computation mainly s3.0 family of tools. 3, Fixed dox-doc. 69 * 70 * Revision 1.1.2.1 2005/07/20 19:37:09 arthchan2003 71 * Added a multi-stream cont_mgau (ms_mgau) which is a wrapper of both gauden and senone. Add ms_mgau_init and model_set_mllr. This allow eliminating 600 lines of code in decode_anytopo/align/allphone. 72 * 73 * 74 * 75 */ 76 77 /** \file ms_mgau.h 78 * 79 * \brief (Sphinx 3.0 specific) A module that wraps up the code of 80 * gauden and senone because they are closely related. 81 * 82 * At the time at Sphinx 3.1 to 3.2, Ravi has decided to rewrite only 83 * single-stream part of the code into cont_mgau.[ch]. This marks the 84 * beginning of historical problem of having two sets of Gaussian 85 * distribution computation routine, one for single-stream and one of 86 * multi-stream. 87 * 88 * In Sphinx 3.5, when we figure out that it is possible to allow both 89 * 3.0 family of tools and 3.x family of tools to coexist. This 90 * becomes one problem we found that very hard to reconcile. That is 91 * why we currently allow two versions of the code in the code 92 * base. This is likely to change in the future. 93 */ 94 95 96 #ifndef _LIBFBS_MS_CONT_MGAU_H_ 97 #define _LIBFBS_MS_CONT_MGAU_H_ 98 99 #include <prim_type.h> 100 #include <cmd_ln.h> 101 #include <logmath.h> 102 #include <s3types.h> 103 #include <ms_gauden.h> 104 #include <ms_senone.h> 105 #include <interp.h> 106 #include <feat.h> 107 #include <mdef.h> 108 #include <ascr.h> 109 110 #ifdef __cplusplus 111 extern "C" { 112 #endif 113 #if 0 114 /* Fool Emacs. */ 115 } 116 #endif 117 118 /* Lists of senones sharing each mixture Gaussian codebook */ 119 /* \struct mgau2sen_t 120 \brief a mapping from gaussian to senone 121 */ 122 typedef struct mgau2sen_s { 123 s3senid_t sen; /**< Senone shared by this mixture Gaussian */ 124 struct mgau2sen_s *next; /**< Next entry in list for this mixture Gaussian */ 125 } mgau2sen_t; 126 127 /** \struct ms_mgau_t 128 \brief Multi-stream mixture gaussian. It is not necessary to be continr 129 */ 130 131 typedef struct { 132 gauden_t* g; /**< The codebook */ 133 senone_t* s; /**< The senone */ 134 mgau2sen_t **mgau2sen; /**< Senones sharing mixture Gaussian codebooks */ 135 interp_t* i; /**< The interplotion weight file */ 136 int32 topn; /**< Top-n gaussian will be computed */ 137 138 /**< Intermediate used in computation */ 139 gauden_dist_t ***dist; 140 int8 *mgau_active; 141 142 } ms_mgau_model_t; 143 144 #define ms_mgau_gauden(msg) (msg->g) 145 #define ms_mgau_senone(msg) (msg->s) 146 #define ms_mgau_interp(msg) (msg->i) 147 #define ms_mgau_mgau2sen(msg) (msg->mgau2sen) 148 #define ms_mgau_topn(msg) (msg->topn) 149 150 ms_mgau_model_t* ms_mgau_init (const char *meanfile, /**< In: File containing means of mixture gaussians */ 151 const char *varfile, /**< In: File containing variances of mixture gaussians */ 152 float64 varfloor, /**< In: Floor value applied to variances; e.g., 0.0001 */ 153 const char *mixwfile, /**< In: File containing mixture weights */ 154 float64 mixwfloor, /**< In: Floor value for mixture weights; e.g., 0.0000001 */ 155 int32 precomp, 156 const char* senmgau, /**< In: type of the gaussians distribution, .cont. or .semi. FIX 157 me! This is confusing!*/ 158 const char* lambdafile, /**< In: Interplation file */ 159 int32 topn, /**< In: Top-n gaussian will be computed */ 160 logmath_t *logmath 161 ); 162 163 /** Free memory allocated by ms_mgau_init */ 164 void ms_mgau_free(ms_mgau_model_t *g /**< In: A set of models to free */ 165 ); 166 167 S3DECODER_EXPORT 168 int32 ms_cont_mgau_frame_eval (ascr_t *ascr, /**< In: An ascr object*/ 169 ms_mgau_model_t *msg, /**< In: A multi-stream mgau mode */ 170 mdef_t *mdef, /**< In: A mdef */ 171 float32** feat 172 ); 173 174 175 S3DECODER_EXPORT 176 int32 model_set_mllr(ms_mgau_model_t* msg, /**< The model-stream Gaussian distribution model */ 177 const char *mllrfile, /**< The MLLR file name */ 178 const char *cb2mllrfile, /**< The codebook to MLLR file name */ 179 feat_t* fcb, /**< FCB object */ 180 mdef_t *mdef, /**< A model definition */ 181 cmd_ln_t *config 182 ); 183 184 #ifdef __cplusplus 185 } 186 #endif 187 188 189 #endif /* _LIBFBS_MS_CONT_MGAU_H_*/ 190 191