cut-the-crap-1.4.0: includes/speech_recognition.c

#include "speech_recognition.h"
#include <pocketsphinx.h>

detect_result* detect_words(char* filepath){
    printf("\n\n\nusing from c %s, \n", filepath);
    printf("size of: (worduh framuh %lu), (word_frame* %lu), (int %lu), (detect_results %lu), (detected_words %lu), (result_code %lu)",
           sizeof(word_frame),
           sizeof(word_frame*),
           sizeof(int),
           sizeof(detect_result),
           sizeof(detected_words),
           sizeof(result_code)
           );
    cmd_ln_t* config = cmd_ln_init(NULL, ps_args(), TRUE,
		         "-hmm", MODELDIR "/en-us/en-us",
	                 "-lm", MODELDIR "/en-us/en-us.lm.bin",
	                 "-dict", MODELDIR "/en-us/cmudict-en-us.dict",
	                 NULL);

    detect_result* res = malloc(sizeof(detect_result));
    if (config == NULL) {
        res->code = FAILED_CONFIG_OBJECT;
	return res;
    }

    ps_decoder_t* decoder = ps_init(config);
    if (decoder == NULL) {
        res->code = FAILED_CONFIG_OBJECT;
	return res;
    }

    // If you want to create your own version of it: it needs to be a single-channel (monaural), little-endian, unheadered 16-bit signed PCM audio file sampled at 16000 Hz.
    FILE* audioFile = fopen(filepath, "rb");
    if (audioFile == NULL) {
        res->code = FAILED_UNABLE_INPUTFILE;
	return res;
    }
    int rv = ps_start_utt(decoder);
    int16 buf[512];

    // do the actuall speech recongnition
    while (!feof(audioFile)) {
	size_t nsamp;
	nsamp = fread(buf, 2, 512, audioFile);
	rv = ps_process_raw(decoder, buf, nsamp, FALSE, FALSE);
        printf("cur rv: %i - %i \n", rv, (int) nsamp);
    }
    rv = ps_end_utt(decoder); // make decoder 'realize' recognintion is over

    word_frame* result = (word_frame*) malloc(5*sizeof(word_frame));
    size_t used = 0;
    size_t size = 5;
    for(ps_seg_t* iterator = ps_seg_iter(decoder); iterator != NULL; iterator = ps_seg_next(iterator)){
        if(used >= size){
          size *= 2;
          result = (word_frame*) realloc(result, size * sizeof(word_frame));
        }
        word_frame current_word_frame;

        ps_seg_frames(iterator,
                      &current_word_frame.from_frame,
                      &current_word_frame.to_frame
                      );
        
        char const* word = ps_seg_word(iterator);

        current_word_frame.word = (char*) malloc(strlen(word)*sizeof(char));
        strcpy(current_word_frame.word, word);

        result[used] = current_word_frame;
        used++;
    }

    fclose(audioFile);
    ps_free(decoder);
    cmd_ln_free_r(config);

    detected_words somevar = {
            result,
            used
    };
    res->code = SUCCESS;
    res->words = somevar;
    return res;
}