/**
* Convert string of space-separated word strings to array of word ids.
*
* @param winfo [in] word dictionary
* @param s [in] string of space-separated word strings
* @param len_return [out] number of found words
*
* @return pointer to a newly allocated word list.
*/
WORD_ID *
new_str2wordseq(WORD_INFO *winfo, char *s, int *len_return)
{
char *p;
int num;
int maxnum;
WORD_ID *wseq;
maxnum = WSSTEP;
wseq = (WORD_ID *)mymalloc(sizeof(WORD_ID)*maxnum);
num = 0;
for (p = strtok(s, " "); p != NULL; p = strtok(NULL, " ")) {
if (num >= maxnum) {
maxnum += WSSTEP;
wseq = (WORD_ID *)myrealloc(wseq, sizeof(WORD_ID) * maxnum);
}
if ((wseq[num] = voca_lookup_wid(p, winfo)) == WORD_INVALID) {
/* not found */
jlog("Error: voca_lookup: word \"%s\" not found in dict\n", p);
free(wseq);
return NULL;
}
num++;
}
*len_return = num;
return(wseq);
}
/**
* <JA>
* @brief 単語辞書をファイルから読み込んでセットアップする.
*
* 辞書上のモノフォン表記からトライフォンへの計算は init_voca() で
* 読み込み時に行われる. このため，辞書読み込み時には，認識で使用する
* 予定のHMM情報を与える必要がある.
*
* N-gram 使用時は，文頭無音単語およぶ文末無音単語をここで設定する.
* また，"-iwspword" 指定時は，ポーズ単語を辞書の最後に挿入する.
*
* </JA>
* <EN>
* @brief Read in word dictionary from a file and setup for recognition.
*
* Monophone-to-triphone conversion will be performed inside init_voca().
* So, an HMM definition data that will be used with the LM should also be
* specified as an argument.
*
* When reading dictionary for N-gram, sentence head silence word and
* tail silence word will be determined in this function. Also,
* when an option "-iwspword" is specified, this will insert a pause
* word at the last of the given dictionary.
*
* </EN>
*
* @param lmconf [in] LM configuration variables
* @param hmminfo [in] HMM definition of each phone in dictionary, for
* phone checking and monophone-to-triphone conversion.
*
* @return the newly created word dictionary structure, or NULL on failure.
*
*/
static WORD_INFO *
initialize_dict(JCONF_LM *lmconf, HTK_HMM_INFO *hmminfo)
{
WORD_INFO *winfo;
JCONF_LM_NAMELIST *nl;
char buf[MAXLINELEN];
int n;
/* allocate new word dictionary */
winfo = word_info_new();
/* read in dictinary from file */
if ( !
#ifdef MONOTREE
/* leave winfo monophone for 1st pass lexicon tree */
init_voca(winfo, lmconf->dictfilename, hmminfo, TRUE, lmconf->forcedict_flag)
#else
init_voca(winfo, lmconf->dictfilename, hmminfo, FALSE, lmconf->forcedict_flag)
#endif
) {
jlog("ERROR: m_fusion: failed to read dictionary, terminated\n");
word_info_free(winfo);
return NULL;
}
/* load additional entries */
for (nl = lmconf->additional_dict_files; nl; nl=nl->next) {
FILE *fp;
if ((fp = fopen(nl->name, "rb")) == NULL) {
jlog("ERROR: m_fusion: failed to open %s\n",nl->name);
word_info_free(winfo);
return NULL;
}
n = winfo->num;
while (getl_fp(buf, MAXLINELEN, fp) != NULL) {
if (voca_load_line(buf, winfo, hmminfo) == FALSE) break;
}
if (voca_load_end(winfo) == FALSE) {
if (lmconf->forcedict_flag) {
jlog("Warning: m_fusion: the error words above are ignored\n");
} else {
jlog("ERROR: m_fusion: error in reading dictionary %s\n", nl->name);
fclose(fp);
word_info_free(winfo);
return NULL;
}
}
if (fclose(fp) == -1) {
jlog("ERROR: m_fusion: failed to close %s\n", nl->name);
word_info_free(winfo);
return NULL;
}
jlog("STAT: + additional dictionary: %s (%d words)\n", nl->name, winfo->num - n);
}
n = winfo->num;
for (nl = lmconf->additional_dict_entries; nl; nl=nl->next) {
if (voca_load_line(nl->name, winfo, hmminfo) == FALSE) {
jlog("ERROR: m_fusion: failed to set entry: %s\n", nl->name);
}
}
if (lmconf->additional_dict_entries) {
if (voca_load_end(winfo) == FALSE) {
jlog("ERROR: m_fusion: failed to read additinoal word entry\n");
word_info_free(winfo);
return NULL;
}
jlog("STAT: + additional entries: %d words\n", winfo->num - n);
}
if (lmconf->lmtype == LM_PROB) {
/* if necessary, append a IW-sp word to the dict if "-iwspword" specified */
if (lmconf->enable_iwspword) {
if (
#ifdef MONOTREE
voca_append_htkdict(lmconf->iwspentry, winfo, hmminfo, TRUE)
#else
voca_append_htkdict(lmconf->iwspentry, winfo, hmminfo, FALSE)
#endif
== FALSE) {
jlog("ERROR: m_fusion: failed to make IW-sp word entry \"%s\"\n", lmconf->iwspentry);
word_info_free(winfo);
return NULL;
} else {
jlog("STAT: 1 IW-sp word entry added\n");
}
}
/* set {head,tail}_silwid */
winfo->head_silwid = voca_lookup_wid(lmconf->head_silname, winfo);
if (winfo->head_silwid == WORD_INVALID) { /* not exist */
jlog("ERROR: m_fusion: head sil word \"%s\" not exist in voca\n", lmconf->head_silname);
word_info_free(winfo);
return NULL;
}
winfo->tail_silwid = voca_lookup_wid(lmconf->tail_silname, winfo);
if (winfo->tail_silwid == WORD_INVALID) { /* not exist */
jlog("ERROR: m_fusion: tail sil word \"%s\" not exist in voca\n", lmconf->tail_silname);
word_info_free(winfo);
return NULL;
}
}
return(winfo);
}
