[asterisk-speech-rec] Grammar not loading...

[Steve Mathers]

Change /etc/lumenvox/sre_server.conf

Very first line:

HMM_TYPE = CONT

To

HMM_TYPE = SEMI

Reload lumenvox..

The continuous (CONT) technology does not support anything but English at the moment. I had to change this to get my Spanish grammars working.

-----Original Message-----
From: asterisk-speech-rec-bounces at lists.digium.com [mailto:asterisk-speech-rec-bounces at lists.digium.com] On Behalf Of Carlos Chavez
Sent: Monday, March 01, 2010 5:14 PM
To: asterisk-speech-rec at lists.digium.com
Subject: [asterisk-speech-rec] Grammar not loading...

        I am having a problem trying to load a grammar in any language other
than en-US.  Here is my dialplan:

[ Context 'lumenvox-test' created by 'pbx_config' ]
  's' =>            1. Answer()       [pbx_config]
                    2. Wait(1)        [pbx_config]
                    3. SpeechCreate()     [pbx_config]
                    4.
SpeechLoadGrammar(estados,/etc/asterisk/grammars/estados.gram)
[pbx_config]
                    5. SpeechActivateGrammar(estados)   [pbx_config]
                    6. SpeechBackground(beep)          [pbx_config]
                    7. Verbose(1,Result was ${SPEECH_TEXT(0)})
[pbx_config]
                    8. Verbose(1,Confidence was ${SPEECH_SCORE(0)})
[pbx_config]
                    9. SpeechDestroy()      [pbx_config]
                    10. Congestion()                [pbx_config]

-= 1 extension (10 priorities) in 1 context. =-

Here is the grammar file:

[root at pbxoficina grammars]# cat estados.gram
#ABNF 1.0 UTF-8;
language es-MX;
mode voice;

root $estados;

$ags = Aguascalientes;
$bc = Baja California;
$bcs = Baja California Sur;
$camp = Campeche;
$df = DF | Distrito Federal;

$estados = $ags | $bc | $bcs | $camp | $df;

        When I try to dial this extension I always get this error:

    -- Executing [2222 at oficina:1] Goto("SIP/2001-00000060",
"lumenvox-test,s,1") in new stack
    -- Goto (lumenvox-test,s,1)
    -- Executing [s at lumenvox-test:1] Answer("SIP/2001-00000060", "") in
new stack
    -- Executing [s at lumenvox-test:2] Wait("SIP/2001-00000060", "1") in
new stack
    -- Executing [s at lumenvox-test:3] SpeechCreate("SIP/2001-00000060",
"") in new stack
    -- Executing [s at lumenvox-test:4]
SpeechLoadGrammar("SIP/2001-00000060",
"estados,/etc/asterisk/grammars/estados.gram") in new stack
[Mar  1 18:08:46] ERROR[23893]: res_speech_lumenvox.c:582 lumenvox_load:
Failed to load grammar 'estados': Couldn't find the grammar
  == Spawn extension (lumenvox-test, s, 4) exited non-zero on
'SIP/2001-00000060'

        If I simply change the language line to en-US the grammar loads but
obviously nothing gets recognized because of the language settings.  I
can see that the Mexican Spanish dictionary is installed in my system:

/etc/lumenvox/Lang/Dict/OtherLanguages/es-MX
/etc/lumenvox/Lang/Dict/OtherLanguages/es-MX/MexicanSpanish.model
/etc/lumenvox/Lang/Dict/OtherLanguages/es-MX/MexicanSpanishDigits.model

        Why is it failing to find any language other than en-US?  I have tried
en-GB and en-AU and they also fail to load.  Am I missing a variable
somewhere to locate other languages?  My grammar files used to work with
LumenVox 8.X but since upgrading to 9.X they do not.

--
Telecomunicaciones Abiertas de México S.A. de C.V.
Carlos Chávez Prats
Director de Tecnología
+52-55-91169161 ext 2001


__________ Information from ESET NOD32 Antivirus, version of virus signature database 4906 (20100301) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com



__________ Information from ESET NOD32 Antivirus, version of virus signature database 4909 (20100302) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com