Nexiwave Asterisk Integration Guide

Have You Considered?

Nexiwave's "Voicemail-to-text Email Service" does not require any coding. Simply configure Asterisk to send voicemail audio to a Nexiwave email address. Our automated service will receive, transcribe and send the Voicemail Transcription email to your destination address. You save valuable development and testing time and instant enterprise class service integration.

Follow these simple steps to add Voicemail-to-text / Voicemail Transcription to your Asterisk or Asterisk based server.

Pre-requisites:

• External applications required: dos2unix, unix2dos, base64, wget, lame(required for mp3). Install them with:
  yum -y install dos2unix unix2dos base64 wget lame
• Asterisk must have been configured to send voicemail emails, as wav attachment.

Steps:

1. Save this script nexiwave_v2t_mail.sh to /etc/asterisk/
2. Make the script executable:
   
   chmod +x /etc/asterisk/nexiwave_v2t_mail.sh
3. Open /etc/asterisk/voicemail.conf. Insert/replace this line in the [general] section (replace Nexiwave username/password with proper values):
   mailcmd=/etc/asterisk/nexiwave_v2t_mail.sh <nexiwave username> <nexiwave password>
4. Restart asterisk:
   service asterisk restart

Your users can now enjoy the great time saving of Voicemail Transcription.

Troubleshooting

To troubleshoot, turn on debug mode in nexiwave_v2t_mail.sh by using this mailcmd in asterisk:

After restart asterisk, test with some voicemails. The script logs debug messages in this file: /tmp/nexiwave_v2t_mail.log. Please examine the log file for errors. 

Debugging tips:

• In debug mode, temporary email files are left undeleted in a temp folder. These files can be very useful for debugging. The location of the temp folder can be found in the log file /tmp/nexiwave_v2t_mail.log.
• Remember to turn off debug mode to prevent excess use of disk space.

Notes:

• This script appends transcript to the end of the email composed by Asterisk. You may want to customize this. To customize, edit the script in section commented by "Customize mail content here"
• For best accuracy, Nexiwave recommends sending un-compressed wav file for transcription. Modify /etc/asterisk/voicemail.conf:
  format=wav
• Redhat/CentOS flavor of Linux is supported. However, the script is simple enough and can be easily modified for other OSes.