Table of Contents

Lumenvox ASR

About

LumenVox LLC
3615 Kearny Villa Road #202
San Diego, CA 92123
USA

Website: http://www.lumenvox.com

Definition

The LumenVox Speech Engine is the only speech software that is directly programmed to work with the connector bridge and will allow your Asterisk applications to accept input via a caller's voice. LumenVox supports US, UK and Australian English, Mexican and South American Spanish, and Canadian French.

Languages

English (UK, US) · French (FR, Canadian) · Spanish (North America)

Installation

Before configuring speech recognition for the VoiceXML browser, install the LumenVox Speech Engine. Complete instructions for downloading and installing LumenVox on Asterisk can be found here: http://www.lumenvox.com/help/speechEngineAsterisk/index.htm.

Following are the main steps for a Debian distribution installation:

Add in /etc/apt/sources.list

deb http://www.lumenvox.com/packages/Debian binary/

Install the following packages:

apt-get install lumenvoxcore
apt-get install lumenvoxclient
apt-get install lumenvoxsre
apt-get install lumenvoxlicenseserver

Start LumenVox deamons.

/etc/init.d/lvlicensed start
/etc/init.d/lvsred start

Generate the Server ID File.

/usr/bin/lv_license_manager -g /root/Info.bts

Upload it in the LumenVox server with your account, and download the license.

Install the license.

/usr/bin/lv_license_manager –m /root/Licensexxxx_xx.bts

Download and install the Asterisk 1.4 connector bridge (tar.gz).

cp res_speech_lumenvox.so /usr/lib/asterisk/modules/
cp lumenvox.conf /etc/asterisk/

Lastly, restart Asterisk.

Configuration

To enable speech recognition, change the main speech parameter.

The “speech” parameter can get three values, “yes”, “automatic”, “no” or “emulation” (don’t generate errors if you enable speech grammars).

[general]
…
speech=automatic
speechprovider=lumenvox
…


You can specify the path directory where the built in grammars are stored with the “speechdirectory” parameter. The default value is “/opt/lumenvox/engine/vxi/%s/%s.gram.”

With the %s markers you can specify two parameters:

//* // Language parameters (examples: “es-ES” or “en-UK”)
//* // Grammar name type, from the field attribute, “type” (examples: “digits”, “number”)

[general]
…
speechdirectory=/opt/lumenvox/engine/vxi/%s/%s.gram
…


In the VoiceXML browser configuration file :

############################
# ASR server configuration #
############################

client.rec.resource.0.cacheDir VXIString /tmp/cacheContent
client.rec.resource.0.format VXIString grxml


You should create the grammar files or create symbolic links to the default grammars installed with the LumenVox package.

cd /opt/lumenvox/engine
mkdir vxi
cd vxi
mkdir en-UK
ln -s ../../Lang/BuiltinGrammars/ABNFNumber.gram number.gram
ln -s ../../Lang/BuiltinGrammars/ABNFDigits.gram digits.gram
ln -s ../../Lang/BuiltinGrammars/ABNFPhone.gram phone.gram


The following VoiceXML example uses the speech recognition, with the built in grammar, ‘digits.’

<?xml version="1.0" encoding="iso-8859-1"?>
<vxml version="2.0" xmlns="http://www.w3.org/2001/vxml" xml:lang="en-UK">
 <form>
  <property name="inputmodes" value="voice"/>
  <property name="timeout" value="30s"/>
  <field name="text" type="digits">
   <catch event="noinput nomatch">
    <reprompt/>
   </catch>
   <prompt>
    Speak to me:
   </prompt>
  </field>
  <filled>
   <prompt>
    You say me:
    <value expr="text" />
   </prompt>
   <clear namelist="text" />
  </filled>
 </form>
</vxml>

Release Notes

To enable dynamic grammars you need to configure the Vxi VoiceXML browser to generate temporary grammar files (look at our Vxi Developer Guide).