CMU Sphinx / Forums / Help: SphinxTrain snapshot help

lag - 2012-04-04

Hi, to not add a topic for every question I think it's better to ask all what
I need just here

I'm using SphinxTrain snapshot and when I run "sphinxtrain -t test setup" it
just add the feat.params and sphinx_train.cfg in the etc folder, there is no
other folders (feat logdir model_parameters model_architecture...) just etc
and wav that I've creat before

and inside of sphinx_train.cfg I've this:

# Variables used in characterizing models $CFG_HMM_TYPE = '.cont.'; # Sphinx III #$CFG_HMM_TYPE = '.semi.'; # PocketSphinx and Sphinx II #$CFG_HMM_TYPE = '.ptm.'; # PocketSphinx (larger data sets)

but in the tutorial:

$CFG_HMM_TYPE = '.cont.'; # Sphinx 4, Pocketsphinx #$CFG_HMM_TYPE = '.semi.'; # PocketSphinx #$CFG_HMM_TYPE = '.ptm.'; # PocketSphinx (larger data sets)

and I just wanna make sure I did the right thing with the .lm.DMP, I did what
in the language model tuto

sphinx_lm_convert -i model.lm -o model.dmp sphinx_lm_convert -i model.dmp -ifmt dmp -o model.lm -ofmt arpa

, after that I renamed the .dmp to .lm.DMP, am I right?!

last question: I based my work on the an4 folder, I did the same
wav/amdigits_clstk and wav/amdigitstest_clstk for the test, my suestion is the
test data should I use some data from the amdigits_clstk or something else...

Thank you so much for all the help
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

lag - 2012-04-04

Is there anybody here??!!!

I want to build an application with pocketsphinx, can I use code::blocks and
Qt to creat an interface?

and what I've to import from sphinxbase and pocketsphinx, can you please
explain, I'm totaly lost :s

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Nickolay V. Shmyrev - 2012-04-07

but in the tutorial: $CFG_HMM_TYPE = '.cont.'; # Sphinx 4, Pocketsphinx

This documentation difference was fixed in trunk

and I just wanna make sure I did the right thing with the .lm.DMP, I did
what in the language model tuto sphinx_lm_convert -i model.lm -o model.dmp
sphinx_lm_convert -i model.dmp -ifmt dmp -o model.lm -ofmt arpa, after that I
renamed the .dmp to .lm.DMP, am I right?!

Only first command is enough. Second command is needed to convert model back.

There was no need to rename extension to uppercase

last question: I based my work on the an4 folder, I did the same
wav/amdigits_clstk and wav/amdigitstest_clstk for the test, my suestion is the
test data should I use some data from the amdigits_clstk or something else...

For the test data you need to use recordings closer to the audio you are going
to recognize, not an4.

I want to build an application with pocketsphinx, can I use code::blocks and
Qt to creat an interface?

Yes you can

and what I've to import from sphinxbase and pocketsphinx, can you please
explain, I'm totaly lost :s

See http://forums.codeblocks.org/index.php?topic=9672.0

http://wiki.codeblocks.org/index.php?title=FAQ-Compiling_%28general%29#Q:My
project_should_be_compiled_with_a_custom_makefile._Is_it_possible_with_Code::B
locks.3F

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Nickolay V. Shmyrev - 2012-04-07

Also note that you do not need to import sphinxbase in order to develop
applications using it. You can just add libraries and includes

http://www.learncpp.com/cpp-tutorial/a3-using-libraries-with-
codeblocks/

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Thank you much sir :) :) :)

Yes I know how to import .h/c/cpp files to my project, but I just don't know
the right ones to import, can you please tell me the right one to import in
order to build a cpp application with a simple qt interface, that records
voice from mic and compare it with my acoustic model... May be using this
function:

static void
recognize_from_microphone()
{
    ad_rec_t *ad;
    int16 adbuf[4096];
    int32 k, ts, rem;
    char const *hyp;
    char const *uttid;
    cont_ad_t *cont;
    char word[256];

    if ((ad = ad_open_dev(cmd_ln_str_r(config, "-adcdev"),
                          (int)cmd_ln_float32_r(config, "-samprate"))) == NULL)
        E_FATAL("Failed top open audio device\n");

    /* Initialize continuous listening module */
    if ((cont = cont_ad_init(ad, ad_read)) == NULL)
        E_FATAL("Failed to initialize voice activity detection\n");
    if (ad_start_rec(ad) < 0)
        E_FATAL("Failed to start recording\n");
    if (cont_ad_calib(cont) < 0)
        E_FATAL("Failed to calibrate voice activity detection\n");

    for (;;) {
        /* Indicate listening for next utterance */
        printf("READY....\n");
        fflush(stdout);
        fflush(stderr);

        /* Wait data for next utterance */
        while ((k = cont_ad_read(cont, adbuf, 4096)) == 0)
            sleep_msec(100);

        if (k < 0)
            E_FATAL("Failed to read audio\n");

        /*
         * Non-zero amount of data received; start recognition of new utterance.
         * NULL argument to uttproc_begin_utt => automatic generation of utterance-id.
         */
        if (ps_start_utt(ps, NULL) < 0)
            E_FATAL("Failed to start utterance\n");
        ps_process_raw(ps, adbuf, k, FALSE, FALSE);
        printf("Listening...\n");
        fflush(stdout);

        /* Note timestamp for this first block of data */
        ts = cont->read_ts;

        /* Decode utterance until end (marked by a "long" silence, >1sec) */
        for (;;) {
            /* Read non-silence audio data, if any, from continuous listening module */
            if ((k = cont_ad_read(cont, adbuf, 4096)) < 0)
                E_FATAL("Failed to read audio\n");
            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;
            }

            /*
             * Decode whatever data was read above.
             */
            rem = ps_process_raw(ps, adbuf, k, FALSE, FALSE);

            /* If no work to be done, sleep a bit */
            if ((rem == 0) && (k == 0))
                sleep_msec(20);
        }

        /*
         * Utterance ended; flush any accumulated, unprocessed A/D data and stop
         * listening until current utterance completely decoded
         */
        ad_stop_rec(ad);
        while (ad_read(ad, adbuf, 4096) >= 0);
        cont_ad_reset(cont);

        printf("Stopped listening, please wait...\n");
        fflush(stdout);
        /* Finish decoding, obtain and print result */
        ps_end_utt(ps);
        hyp = ps_get_hyp(ps, NULL, &uttid);
        printf("%s: %s\n", uttid, hyp);
        fflush(stdout);

        /* Exit if the first word spoken was GOODBYE */
        if (hyp) {
            sscanf(hyp, "%s", word);
            if (strcmp(word, "goodbye") == 0)
                break;
        }

        /* Resume A/D recording for next utterance */
        if (ad_start_rec(ad) < 0)
            E_FATAL("Failed to start recording\n");
    }

    cont_ad_close(cont);
    ad_close(ad);
}

with:

....
config = cmd_ln_init(NULL, ps_args(), TRUE,
                 "-hmm", MODELDIR "/hmm/en_US/hub4wsj_sc_8k",
                 "-lm", MODELDIR "/lm/en/turtle.DMP",
                 "-dict", MODELDIR "/lm/en/turtle.dic",
                 NULL);

Thank's again :)

Nickolay V. Shmyrev - 2012-04-08

Please read the tutorial

http://cmusphinx.sourceforge.net/wiki/tutorialpocketsphinx

It describes which files to import

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

lag - 2012-04-08

Sorry I think I didn't make myself clear, I just want to add this part :

config = cmd_ln_init(NULL, ps_args(), TRUE, "-hmm", MODELDIR "/hmm/en_US/hub4wsj_sc_8k", "-lm", MODELDIR "/lm/en/turtle.DMP", "-dict", MODELDIR "/lm/en/turtle.dic", NULL);

to the pocketsphinx/src/programs/continuous.c, I think there is a function
that allows puting c code in a cpp project ( extern "C" {} I think)

I just need a point to start from, I'm lost right now :/

Sorry
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

lag - 2012-04-08

Hello all,
I did exactely as You said:

" Only first command is enough. Second command is needed to convert model
back. There was no need to rename extension to uppercase

but now I've some errors :/ :

ERROR: "ngram_model_arpa.c", line 466: File /home/hyt-med/Desktop/beta/etc/beta.lm.DMP not found ERROR: "ngram_model_dmp.c", line 106: Dump file /home/hyt-med/Desktop/beta/etc/beta.lm.DMP not found ERROR: "ngram_search.c", line 208: Failed to read language model file: /home/hyt-med/Desktop/beta/etc/beta.lm.DMP

the command sphinx_lm_convert generate a .dmp file but in the tutorial
:(your_db.lm.DMP), I think I've to rename the file from beta.dmp to
beta.lm.DMP so it can be found, Yes?!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

SphinxTrain snapshot help

Speech Recognition Toolkit

Forums

Help

SphinxTrain snapshot help document.SUBSCRIPTION_OPTIONS = { "thing": "topic", "subscribed": false, "url": "subscribe", "icon": { "css": "fa fa-envelope-o" } };

SphinxTrain snapshot help