Get Started with Voice Alerts Using Ruby

    Overview

    This guide shows how to make voice calls to alert customers to critical issues that require immediate attention. You can play recorded audio when the call recipient answers or use text-to-speech. You can then take action based on a dialpad key they press in response. You can set different actions if the call is not answered, if the line is busy, or if you reach voicemail.

    You can send voice alerts either by using our PHLO visual workflow builder or our APIs and XML documents. Follow the instructions in one of the tabs below.

    To implement voice alerts use-case, you can create and deploy a PHLO with a few clicks on the PHLO canvas. PHLO also lets you visually construct your entire use-case. With PHLO, you only pay for calls you make/receive, and building with PHLO is free.

    Outline

    Implementation

    In this section, we will guide you to create a PHLO to implement voice alerts use-case.

    Prerequisites

    1. Create a Plivo Account(if you don’t have one already): You can sign up with your work email address and complete the phone verification step using your mobile number.
    2. Set up your Dotnet dev environment: To set up your dev environment in your .NET(C#) programming language, please refer to the instructions available in the Set up Your Dotnet Dev Environment section.
    3. Buy a Plivo number(optional): You must have a voice-enabled Plivo phone number if you are willing to receive incoming calls. You can purchase numbers from the Numbers section of your Plivo Console. It is also possible to purchase numbers using the Numbers API.

    Create the PHLO

    1. On the top navigation bar, click PHLO. The PHLO page will appear and display your existing PHLOs, if any exist. If this is your first PHLO, then the PHLO page will be empty.
    2. Click CREATE NEW PHLO to build a new PHLO.
    3. In the Choose your use-case window, click Build my own. The PHLO canvas will appear with the Start node. Note: The Start node is the starting point of any PHLO. You can choose between the four available trigger states of the Start node; Incoming SMS, Incoming Call, and API Request. For this PHLO, we will use the API Request trigger state.
    4. Click the Start node to open the Configurations tab, and then enter the information to retrieve from the HTTP Request payload. For example, this could be a list of numbers to call.
    5. Click Validate to save the configurations for the node.
    6. From the list of components on the left-hand side, drag and drop the Initiate Call component onto the canvas. This will add a Initiate call node onto the canvas.
    7. Connect the Start node with the Initiate Call node, using the API Request trigger state.
    8. Configure the Initiate Call node to select the From and To numbers from the Start node. Note:
      • Enter two curly brackets to view all available variables.
      • The values for the From and To numbers are picked from the HTTP Request Payload. Make sure to clearly define the variables in the Start node.
    9. Click Validate to save the configurations for the node.
    10. Similarly, create a node for the IVR Menu component and connect it to the Initiate Call node using the Answered trigger state.
    11. Click the IVR Menu node to open the Configurations tab.
    12. Configure the choices for the IVR Menu from the Configurations tab. In this example, we will select 1 and 2 as the allowed choices and enter a message to play to the user.
    13. Rename the IVR Menu node to Gather_Input. Note: You can rename the nodes as per your requirements. We are using specific names to help you relate to the different nodes used in this use case.
    14. Enter a message in the Speak Text field. The caller will be greeted with this message. Note: You can also configure the Language and Voice for the message.
    15. Click Validate to to save the configurations of the node.
    16. Similarly, create three nodes for the Play Audio component. Note: In this example, we will need three Play Audio nodes; one node for each option (Wrong Input, 1, and 2).
    17. Rename these nodes to Invalid_Input_Prompt, Resolved_Prompt, and** Escalation_Prompt** using the Configurations tab.
    18. Connect Invalid_Input_Prompt node to the Gather_Input using the Wrong Input trigger state.
    19. Connect the Prompt Completed trigger state of Invalid_Input_Prompt node to the IVR Menu. This will send the user back to the IVR Menu if they press an incorrect option, or if they do not press any key.
    20. Connect Resolved_Prompt and Escalation_Prompt nodes to the Gather_Input using the 1 and 2 trigger states.
    21. Next, configure the Play Audio nodes to play a specific message to the user. Note: The audio playback can either be static or dynamic. In this example, we will use a combination of a static message along with some dynamic message for the Resolved_Prompt and Escalation_Prompt nodes.
    22. On the Play Audio Configurations tab, enter a static message (For example, Your Database status is: ) in the Speak Text field, followed by a variable to include the dynamic text. Note: You can choose a variable by entering two curly brackets in the Speak Text field.
    23. Click Validate to save the configurations for the node.
    24. Configure** Resolved_Prompt** and Escalation_Prompt nodes to speak a fixed message in case of a wrong input or no input.
    25. Click Validate to save the configurations for the node.
    26. Next, create a node for Initiate Call component and rename it to Escalation_Call.
    27. Connect the Escalation_Prompt node to the Escalation_Call node using the Prompt Completed trigger state. This will call a second phone number and announce the alert. You can set up any number of escalation numbers by creating similar nodes for each phone number.
    28. Connect the Escalation_Call node back to the Escalation_Prompt node using the Answered trigger state. This will play the same prompt to the second user and the process will repeat.
    29. After you complete the configurations, click Save. You can trigger your PHLO using the PHLO URL to test it out. Your complete PHLO will look like this:

    Voice Alert

    Set up Your Ruby Dev Environment

    You must set up and install Ruby and Plivo’s Ruby SDK to implement voice alerts use-case. Here’s how.

    Install Ruby

    Operating SystemInstructions
    macOS & LinuxYou would already have Ruby installed, you can check this by running the command ruby --version in the terminal. If you don't have it installed, you can install it using homebrew.
    WindowsTo install Ruby on Windows you can download it from here and install.

    Install Plivo Ruby Package

    • Create a project directory, run the following command:

      $ mkdir myrubyapp
      
    • Change the directory to our project directory in the command line:

      $ cd myrubyapp
      
    • Add this line to your application’s Gemfile:

      gem 'plivo', '>= 4.3.0'
      
    • And then execute:

      $ bundle
      
    • Or install it yourself as:

      $ gem install plivo
      

    Trigger the PHLO

    Once you have created and setup your Ruby dev envrironment, you can go to your Plivo Consolse and copy the PHLO_ID. You can integrate a PHLO into your application workflow by making an API request to trigger the PHLO with the required payload.

    With Static Payload

    You can choose to either configure the mandatory params required for a PHLO while creating the PHLO itself or, you can pass the params as payload while triggering the PHLO from your app.

    Note: When you configure the params while creating the PHLO, the values are hardcoded, and they act as static payload.

    With Static Payload

    Code

    Now, create a file called trigger_phlo.rb and paste the below code.

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    
    require 'rubygems'
    require 'plivo'
    
    include Plivo
    
    AUTH_ID = 'AUTH_ID'
    AUTH_TOKEN = 'AUTH_TOKEN'
    
    client = Phlo.new(AUTH_ID, AUTH_TOKEN)
    
    # if credentials are stored in the PLIVO_AUTH_ID and the PLIVO_AUTH_TOKEN environment variables
    # then initialize client as:
    # client = Phlo.new
    
    begin
      phlo = client.phlo.get('phlo_id')
      response = phlo.run()
      puts response
    rescue PlivoRESTError => e
      puts 'Exception: ' + e.message
    end
    

    With Dynamic Payload

    To use dynamic values for the parameters, you can use the liquid templating params while creating the PHLO and pass the values while triggering the PHLO.

    With Dynamic Payload

    Code

    Now, create a file called trigger_phlo.rb and paste the below code.

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    22
    23
    24
    25
    26
    
    require 'rubygems'
    require 'plivo'
    
    include Plivo
    
    AUTH_ID = 'AUTH_ID'
    AUTH_TOKEN = 'AUTH_TOKEN'
    
    client = Phlo.new(AUTH_ID, AUTH_TOKEN)
    
    # if credentials are stored in the PLIVO_AUTH_ID and the PLIVO_AUTH_TOKEN environment variables
    # then initialize client as:
    # client = Phlo.new
    
    begin
      phlo = client.phlo.get('phlo_id')
      #parameters set in PHLO - params
      params = {
         from: '+14157778888',
         to: '+14157778889'
      }
      response = phlo.run(params)
      puts response
    rescue PlivoRESTError => e
      puts 'Exception: ' + e.message
    end
    

    You can get your Auth_ID and Auth_token from your dashboard AUTHID

    You can find the PHLO_ID on the PHLO Listing page. PHLO Listing

    Test and Validate

    You can save the file and use the below command to run your code.

    $ ruby trigger_phlo.rb
    
    Note: If you are using a Plivo Trial account for this example, you can only make calls to phone numbers that have been verified with Plivo. Phone numbers can be verified at the Sandbox Numbers page.
      Here's how to implement voice alerts by using Ruby and the Plivo [Dial](/docs/voice/xml/dial/) XML element handle incoming calls.
    

    First, a little background information. To control call flow, you create a Plivo application and associate it with a Plivo voice-enabled number. The application includes an answer URL.  When a call comes in, Plivo sends an HTTP request to the answer URL with the attributes of the call, and expects a valid XML response back so that it knows how to handle the call.

    Plivo also initiates HTTP requests to your application server through the course of a call based on specific XML elements in your answer XML. Such requests fall broadly into two categories:

    Action URL requests: Plivo expects XML instructions to carry forward the call in response to action URL requests, which are typically invoked at the end of an XML element’s execution, such as when IVR input is received from a caller during GetInput XML execution.

    Callback URL requests: No XML instructions are expected in response to these requests. They serve as webhooks to notify your application server of important events through the course of an XML element’s execution — for example, when a conference participant is muted or unmuted.

    Voice alert workflow

    Outbound- Call Flow

    Prerequisites

    This section will guide you through how to use Plivo APIs to make voice calls from your application. First, let’s make sure you meet these prerequisites before we dive into the code.

    • Plivo Auth Id and Auth Token: You will find your Plivo Auth Id and Auth Token on the home screen of your Plivo Console. Click here to sign-up for a Plivo account if you haven’t already!

      Find Your Auth Credentials on Plivo Console

    • Plivo Phone Number(Optional): You can purchase numbers from the Numbers section of your Plivo Console and use the same as the caller ID for the outbound call. This number will also help you receive incoming calls as you must have a voice-enabled Plivo phone number to do the same. Please note that you can also purchase numbers using the Numbers API.

      Buy a New Plivo Number

    • Answer Url: When a call is answered by the destination_number, you can control the call flow with the help of the answer_url set in the API request. Plivo will invoke the answer_url specified as soon as the call is answered and expect a valid XML response with instructions to handle the call. For example, you can use https://s3.amazonaws.com/static.plivo.com/alert.xml as answer_url to test voice notifications implementation. The XML response of this is:

      <Response> 
      <Speak>Your Database is out of memory! Please press 1 to resolve or 2 to escalate.</Speak> 
      </Response>
      

      As you can tell, the XML above holds instructions to say ‘Your Database is out of memory! Please press 1 to resolve or 2 to escalate.’ to the callee. You can find the entire list of Plivo XML verbs and their details in the XML Reference section of the website here.

    Set up Your Ruby Dev Environment

    You must set up and install Ruby and Plivo’s Ruby SDK to implement voice alerts use-case. Here’s how.

    Install Ruby

    Operating SystemInstructions
    macOS & LinuxYou would already have Ruby installed, you can check this by running the command ruby --version in the terminal. If you don't have it installed, you can install it using homebrew.
    WindowsTo install Ruby on Windows you can download it from here and install.

    Install Plivo Ruby Package

    • Create a project directory, run the following command:

      $ mkdir myrubyapp
      
    • Change the directory to our project directory in the command line:

      $ cd myrubyapp
      
    • Add this line to your application’s Gemfile:

      gem 'plivo', '>= 4.3.0'
      
    • And then execute:

      $ bundle
      
    • Or install it yourself as:

      $ gem install plivo
      

    Create the Voice Alerts App

    Once you have the above prerequisites set, You can follow the below instructions to create an app to make an outbound call and leave a text-to-speech message(as Voice Alert) once the call recipient answers the call.

    Now, create a file called make_call.rb and paste the below code.

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    
    require 'rubygems'
    require 'plivo'
    
    include Plivo
    include Plivo::Exceptions
    
    api = RestClient.new("<auth_id>","<auth_token>")
    
    begin
      response = api.calls.create(
        '+14151234567',
        ['+14157654321'],
        'https://s3.amazonaws.com/static.plivo.com/alert.xml'
      )
      puts response
    rescue PlivoRESTError => e
      puts 'Exception: ' + e.message
    end
    
    Note:
    • Replace the placeholders auth_id & auth_token with your credentials from Plivo Console.
    • We recommend that you store your credentials in the auth_id & auth_token environment variables, so as to avoid the possibility of accidentally committing them to source control. If you do this, you can initialize the client with no arguments and it will automatically fetch them from the environment variables
    • You can use ENV to store environment variables and fetch them while initializing the client.
    • Replace the placeholder +14151234567 with the Phone number which you have purchased and +14157654321 with the phone number you will be making calls to.
    • Both +14151234567 and +14157654321 should be in E.164 format

    Test and Validate

    Once you have created the Voice Alerts App, you can save the file and use the below command to run your code.

    $ ruby make_call.rb
    
    Note: If you are using a Plivo Trial account for this example, you can only make calls to phone numbers that have been verified with Plivo. Phone numbers can be verified at the Sandbox Numbers page.