1 /*************************************************************************/
2 /* */
3 /* Centre for Speech Technology Research */
4 /* University of Edinburgh, UK */
5 /* Copyright (c) 1996,1997 */
6 /* All Rights Reserved. */
7 /* */
8 /* Permission is hereby granted, free of charge, to use and distribute */
9 /* this software and its documentation without restriction, including */
10 /* without limitation the rights to use, copy, modify, merge, publish, */
11 /* distribute, sublicense, and/or sell copies of this work, and to */
12 /* permit persons to whom this work is furnished to do so, subject to */
13 /* the following conditions: */
14 /* 1. The code must retain the above copyright notice, this list of */
15 /* conditions and the following disclaimer. */
16 /* 2. Any modifications must be clearly marked as such. */
17 /* 3. Original authors' names are not deleted. */
18 /* 4. The authors' names are not used to endorse or promote products */
19 /* derived from this software without specific prior written */
20 /* permission. */
21 /* */
22 /* THE UNIVERSITY OF EDINBURGH AND THE CONTRIBUTORS TO THIS WORK */
23 /* DISCLAIM ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING */
24 /* ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT */
25 /* SHALL THE UNIVERSITY OF EDINBURGH NOR THE CONTRIBUTORS BE LIABLE */
26 /* FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES */
27 /* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN */
28 /* AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, */
29 /* ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF */
30 /* THIS SOFTWARE. */
31 /* */
32 /*************************************************************************/
33 /* Author : Alan W Black */
34 /* Date : March 1997 */
35 /*-----------------------------------------------------------------------*/
36 /* */
37 /* This assigns a form of part of speech to tokens. It is typically use */
38 /* to assign gross level pos for things before they are converted to */
39 /* words, particularly, numbers (ordinals, digitized, years, numbers) */
40 /* */
41 /*=======================================================================*/
42 #include <cstdio>
43 #include "festival.h"
44 #include "tokenP.h"
45
FT_Token_POS_Utt(LISP utt)46 LISP FT_Token_POS_Utt(LISP utt)
47 {
48 // This module assigns token_pos feature to each token based on the
49 // Assoc list of Regex to CART trees in token_pos_cart_trees.
50 EST_Utterance *u = get_c_utt(utt);
51 EST_Item *t;
52 LISP trees,l;
53
54 trees = siod_get_lval("token_pos_cart_trees",NULL);
55 if (trees == NIL) return utt;
56
57 for (t=u->relation("Token")->first(); t != 0; t = t->next())
58 {
59 if (t->f("token_pos","0") == "0")
60 for (l=trees; l != NIL; l=cdr(l)) // find a tree that matches
61 {
62 if (t->name().matches(make_regex(get_c_string(car(car(l))))))
63 {
64 t->set_val("token_pos",
65 wagon_predict(t,car(cdr(car(l)))));
66 break;
67 }
68 }
69 }
70
71 return utt;
72
73 }
74