/*
* Continuous recognition from a file
*/
static void
recognize_from_file()
{
int16 adbuf[2048];
const char *fname;
const char *hyp;
int32 k;
uint8 utt_started, in_speech;
int32 print_times = cmd_ln_boolean_r(config, "-time");
fname = cmd_ln_str_r(config, "-infile");
if ((rawfd = fopen(fname, "rb")) == NULL) {
E_FATAL_SYSTEM("Failed to open file '%s' for reading",
fname);
}
if (strlen(fname) > 4 && strcmp(fname + strlen(fname) - 4, ".wav") == 0) {
char waveheader[44];
fread(waveheader, 1, 44, rawfd);
if (!check_wav_header(waveheader, (int)cmd_ln_float32_r(config, "-samprate")))
E_FATAL("Failed to process file '%s' due to format mismatch.\n", fname);
}
if (strlen(fname) > 4 && strcmp(fname + strlen(fname) - 4, ".mp3") == 0) {
E_FATAL("Can not decode mp3 files, convert input file to WAV 16kHz 16-bit mono before decoding.\n");
}
ps_start_utt(ps);
utt_started = FALSE;
while ((k = fread(adbuf, sizeof(int16), 2048, rawfd)) > 0) {
ps_process_raw(ps, adbuf, k, FALSE, FALSE);
in_speech = ps_get_in_speech(ps);
if (in_speech && !utt_started) {
utt_started = TRUE;
}
if (!in_speech && utt_started) {
ps_end_utt(ps);
hyp = ps_get_hyp(ps, NULL);
if (hyp != NULL)
printf("%s\n", hyp);
if (print_times)
print_word_times();
fflush(stdout);
ps_start_utt(ps);
utt_started = FALSE;
}
}
ps_end_utt(ps);
if (utt_started) {
hyp = ps_get_hyp(ps, NULL);
if (hyp != NULL) {
printf("%s\n", hyp);
if (print_times) {
print_word_times();
}
}
}
fclose(rawfd);
}
/*
* Continuous recognition from a file
*/
static void
recognize_from_file() {
cont_ad_t *cont;
ad_rec_t file_ad = {0};
int16 adbuf[4096];
const char* hyp;
const char* uttid;
int32 k, ts, start;
char waveheader[44];
if ((rawfd = fopen(cmd_ln_str_r(config, "-infile"), "rb")) == NULL) {
E_FATAL_SYSTEM("Failed to open file '%s' for reading",
cmd_ln_str_r(config, "-infile"));
}
fread(waveheader, 1, 44, rawfd);
file_ad.sps = (int32)cmd_ln_float32_r(config, "-samprate");
file_ad.bps = sizeof(int16);
if ((cont = cont_ad_init(&file_ad, ad_file_read)) == NULL) {
E_FATAL("Failed to initialize voice activity detection");
}
if (cont_ad_calib(cont) < 0)
E_FATAL("Failed to calibrate voice activity detection\n");
rewind (rawfd);
for (;;) {
while ((k = cont_ad_read(cont, adbuf, 4096)) == 0);
if (k < 0) {
break;
}
if (ps_start_utt(ps, NULL) < 0)
E_FATAL("ps_start_utt() failed\n");
ps_process_raw(ps, adbuf, k, FALSE, FALSE);
ts = cont->read_ts;
start = ((ts - k) * 100.0) / file_ad.sps;
for (;;) {
if ((k = cont_ad_read(cont, adbuf, 4096)) < 0)
break;
if (k == 0) {
/*
* No speech data available; check current timestamp with most recent
* speech to see if more than 1 sec elapsed. If so, end of utterance.
*/
if ((cont->read_ts - ts) > DEFAULT_SAMPLES_PER_SEC)
break;
}
else {
/* New speech data received; note current timestamp */
ts = cont->read_ts;
}
ps_process_raw(ps, adbuf, k, FALSE, FALSE);
}
ps_end_utt(ps);
if (cmd_ln_boolean_r(config, "-time")) {
print_word_times(start);
} else {
hyp = ps_get_hyp(ps, NULL, &uttid);
fprintf(stderr, "%s: %s\n", uttid, hyp);
}
fflush(stdout);
}
cont_ad_close(cont);
fclose(rawfd);
}
