Powering seamless voice interactions using IMIconnect

Voice communication is the most natural way for humans to interact with each other and therefore it’s no surprise that it is also one of the oldest and widely used channels for business-customer interactions. Recent advancements such as biometric authentication via voice, automatic translation and interpretation, and voice-based virtual personal assistants are leading to renewed interest in voice and its application across several new customer engagement scenarios.

According to a recent survey by Microsoft on the ‘Global state of customer service‘, approximately 3 out of 4 respondents across age groups have dialed into a phone line for some form of customer service.

Voice will undeniably remain a key channel for powering customer service interactions even in the future. Here is why:

  • Certain demographics and customer segments will be most comfortable with voice
  • Voice conversations allow you to provide personalized service
  • Customers who dial-in to your IVR usually have an urgent problem
  • Complex problems are easier to diagnose over the phone as compared to messaging
  • Setting up an interactive voice response system is no longer expensive or time-consuming
  • New devices (such as smart watches) are proliferating and form factors are getting smaller. Keyboard input in such devices isn’t always convenient or possible, hence voice becomes a natural choice

How can IMIconnect help you deliver seamless voice experiences?

There is a lot involved in setting up a robust and scalable IVR service if you were to try and do it using traditional approaches. At a minimum, you need to:

  1. Procure PBX or similar systems and connectivity in the form of ISDN PRI lines or SIP lines
  2. Hire an army of developers with very specialized skills
  3. Code your IVR application (menus, conditional branching, integrations etc.)
  4. Test and deploy your application
  5. Ensure sufficient server capacity and multiple lines to support concurrent calls
  6. Redo the exercise ‘Step C’ onwards in case you need to change menu options, etc.

IMIconnect accelerates this entire process and allows you to focus on building the best voice experience as against worrying about the operational challenges.

To begin with, IMIconnect allows you to buy a variety of numbers such as Landline, Multi-functional, or Toll-free numbers from within the platform in just a few clicks.

You can purchase a local number of your choice in 80+ countries is just a few clicks. Once purchased, these numbers are immediately available for building IVR flows.

Voice flow builder

IMIconnect’s voice flow builder allows you to create interactive voice flows with minimal programming or scripting efforts. It has an easy to use drag-and-drop interface called ‘Flow Canvas’ that enables rapid development of multi-step voice flows using pre-built nodes.

The above flow demonstrates a customer authentication journey involving verification of a one-time-password (OTP) using DTMF based keypad input capture. The entire communication logic is built on the flow canvas by connecting and configuring relevant nodes in a logical order. A node contains complete information such as initiation, execution, error handling, condition, notifications, logic – switch statements, and decision logic required to automate complex processes.

As an example, let’s take a quick look at the configuration UI of the ‘Play’ node.

As shown, the ‘Play’ node is used to play one or more voice prompts in the configured sequence. A variety of play options are supported, such as:

  • Play-Prompt: Plays back selected voice file
  • Play-DTMF: Plays back the keys pressed by the user, often used to get a confirmation
  • Play-Number: Converts any number to voice and play it to the caller
  • Play-Currency: Converts a number to currency and plays it to the caller. We currently support 18 currency formats.
  • Play-Date: To read out a date. Custom date formats are supported.
  • Play-Time: Converts a number to time format and plays it to the caller
  • Play-HTTP Prompt: Reads out the configured URL to the caller
  • Play-Stream: Plays a live audio stream from a URL to the caller
  • Text to Speech: Speech synthesis engine. Multiple languages & dialects supported.

IMIconnect voice flows offer you several advanced capabilities to build seamless experiences with end-to-end automation. Here is a quick look at some of the key nodes offered in the voice flow palette:

Node nameDescription
The entry point in the flow
Plays or streams one or more media prompts. Text-to -speech playback is also supported.
Records a voice call and stores it in a media folder on any connected destination.
Transfers an inbound or an outbound call to another destination number.
Collects user input as DTMF keystrokes within the flow. The input may be of fixed length (e.g. 6 digit pins) or variable length followed by a # stroke. DTMF encryption is supported.
Uses JavaScript to execute custom logic and branch the flow into multiple streams based on output.
Calls an external application by passing contextual information to fetch specific data as response. User can branch out the flow to handle exceptions.
Allows invocation of a different voice flow from the existing flow. Helps in modular design of flows where parts of a large IVR flow can be managed as smaller logical constructs.
Supports tone dialling in a voice flow.
Merges one or more active calls and creates a voice conference. User can either create a conference or add a party to an existing conference, set the room size, etc.
Closes an existing conference or removes an active party from the conference call.
Automatic Speech Recognition (ASR) detects keywords from the customer’s speech in a live voice call. Each identified keyword can be used to branch the flow.
Sends message to customers on any of the multiple channels supported on IMIconnect such as SMS, Facebook Messenger, Twitter DM, WhatsApp, RTM, WeChat, Email, Push notifications to Android, iOS, JS SDKs & web browsers
Disconnects call

Building seamless voice experiences using IMIconnect is a simple exercise of setting up logical flows using our drag-and-drop builder. Not just voice, you can easily automate multichannel customer journeys by using voice alongside other channels such as Facebook Messenger, Twitter Direct Messaging, SMS, Email, WeChat, and more. Further, these flows can easily be extended to integrate with backend systems to achieve end-to-end automation.

The underlying infrastructure layer takes care of voice capacity augmentation for managing spikes in voice calls. Whether it is a single customer calling in at a time or 100s of customers reaching out to you simultaneously, the setup process remains the same.

With local numbers available from over 80 countries, and connectivity with all major telecom operators across the globe, the platform is currently powering a variety of voice applications across industry verticals.