Microsoft bot framework Archives - BotCore

Deploy an IVR Bot to Improve Customer Self Service

Abhishek Shanbhag — Mon, 28 Jun 2021 12:11:00 +0000

Deploy an IVR Bot to Improve Customer Self Service

Automated customer interaction is the backbone of any customer-facing communications strategy. Since the advent of the COVID-19 pandemic, there has been a sharp surge in calls to customer service, particularly in industries such as banking, hospitality, insurance, and health.

Moreover, research has shown that nearly 90% of people prefer voice interaction instead of navigating a complex phone menu. However, it can be challenging for organizations to deal with the high volume of incoming customer calls.

For one, contact centers have limited resources to handle the massive influx of calls. Secondly, voice-based channels are not only expensive but difficult to scale. Additionally, many customers communicate via speech but may prefer a self-service channel to convey their requests.

In such a scenario, interactive voice response (IVR) bots can provide an intelligent, engaging, and natural self-service way to support customers – anytime and anywhere.

Let’s explore.

IVR Bot to Improve Customer Self Service

IVR bots are computer programs that can converse like people and simultaneously service an unlimited number of requests, provide 24X7 support, answer customer queries instantly, and decrease business costs.

The bots use artificial intelligence (AI), natural language processing (NLP), and machine learning (ML) technologies to process customer requests, enable them to self-transact, learn from past conversations, provide personalized solutions to their queries, or escalate calls to a human agent when needed.

Customers can get immediate and accurate responses in the language of their choice – it’s fast, scalable, cost-effective, and improves first-call resolution rates.

Therefore, deploying IVR bots infuses agility in customer service operations, allowing agents to focus on the most critical customer requests and tasks that add value to the organization.

Using Microsoft Azure AI to build an automated IVR chatbot

For Microsoft customers looking to build IVR bots, Microsoft Azure AI boasts all crucial elements that help build an automated IVR solution.

Backed by the Microsoft Azure cloud platform, the solution can manage speech requests via Teams, Skype, and the Microsoft Bot Framework with tools for data ingestion, data storage, data processing, and advanced analytics.

Core Azure Services and Microsoft technologies that help build the IVR bot

Microsoft’s Natural Language Processing technology

IVR bots leverage natural language processing (NLP) to understand the nuances of the human language, including grammar, synonyms, and slang.

Speech service (Bing Speech API or Cognitive Services Speech Service)

The Speech service transcribes raw speech data into text form.

Natural Language Understanding (NLU) technology

Language Understanding (LUIS) identifies customer intent and spoken entities from the transcribed text.

Example – If the bot says, “I want to place an order for a grey washing machine.”

Here, “place an order” is the intent, and the “grey washing machine” (product category) is the entity

Microsoft Bot Framework

Manages the call workflow and processes conversation results.

Other supporting technologies

Azure Web app, Azure SQL, Azure Cosmos DB, etc.

Learn More: Interactive voice response app with bot

Essential features of an IVR bot for improved customer self-service

Intent extraction and analysis
Ability to pause, listen, and respond accordingly whenever a customer barges in during an ongoing interaction
Ability to process information and act at the same pace at which the customer speaks
Optimal integration with the contact center database to know the caller history, identify the likely reason for the call, and tailor suggestions based on the customer’s persona.
Ability to hand over the call to a human agent or subject matter expert (SME) when required.
Machine learning capabilities to improve response accuracy and learn from past customer conversations

How can Acuvate help?

As a Microsoft Gold Partner, we at Acuvate helps clients build IVR bots and improve customer self-service with our intuitive, low-code bot-builder platform called BotCore.

The low-code, visual design interface allows the quick deployment of bots.
Seamless integration with legacy enterprise systems and AI services
Multilingual capabilities enable organizations to engage with customers in their native language
We leverage the best of Microsoft technologies, including AI, NLP, Azure Cognitive Services, and LUIS.

To know more about BotCore, please feel free to schedule a personalized consultation with our experts.

The post Deploy an IVR Bot to Improve Customer Self Service appeared first on BotCore.

Why are we excited about Power Virtual Agents?

Abhishek Shanbhag — Mon, 11 Nov 2019 18:30:00 +0000

Why Are We Excited About Power Virtual Agents?

At this year’s Ignite , Microsoft made several announcements related to conversational AI and chatbots. One of the key announcements that caught our attention is the launch of Power Virtual Agents as a part of Microsoft’s citizen development vision of Power Platform.

Power Platform as you know had until now a few powerful tools like Power BI, Power Apps & Flow (now called Power Automate). With the addition of Power Virtual Agents, Microsoft is essentially connecting all their previous investments in these respective technologies. At the same time, addressing the key questions that many customers and partners had been inquisitive about: Where is Microsoft’s bot builder platform?

conversational AI & microsoft

With the launch of the Microsoft Bot Framework in early 2016, Microsoft paved the way for customers and vendors alike to start experimenting and developing chatbots. Simply put, Bot framework is an SDK and coupled with the Azure Bot Service and Cognitive Services like LUIS and QnA Maker using which developers can build some pretty cool chatbots. However, bot building is yet regarded as technical & complex. To bridge this gap and democratize bot building, solutions like BotCore were developed that put the business analyst or the power user in the driver seat.

With Power Virtual Agents rollout, we are comprehending that all these investments that Microsoft has done since 2016 are all coming together. Not to any wonder, Bot Framework and other cognitive services which are used to enable chatbots forms the foundation of Power Virtual Agents. The building experience is however very simple. This platform doesn’t necessitate an understanding of concepts like intents, utterances, entities etc. which every bot developer and trainer swear by. A power business user can go from zero to a working bot in a matter of minutes!

experiments with power virtual agents

One thing that is true with any chatbot rollout is to fail faster and keep trying. Chatbots can be used to automate a lot of different areas. However, if you choose a use case that happens to complicate the end-user experience or make it difficult for a user to get the job done – your chatbot adoption is at stake!

Once you identify a process or area that can be handled by a chatbot, using Power Virtual Agents, you can quickly create, test and deploy the agent to your users. Call it a PoC or MVP or Pilot or anything else that assures your users that there will be more improvements going forward. This helps you to test the waters before embarking on a larger enterprise-wide journey that would take more time and effort.

areas of employee support within the enterprise

Power Virtual Agents platform was developed as part of the Dynamics family of products and was primarily built for customer support use-case. However, the chatbots built using it can be deployed for your internal customers or employees as well. The whole notion behind chatbots is to offer quick front office or L1 support, effectively eliminating or reducing low-value requests that come into your contact centre or helpdesk.

Few of the areas, where we see Power Virtual Agents being used within the enterprise are:

IT Support
- FAQ
- Self-help and troubleshooting
- Triaging a problem before creating an incident or request in ITSM tools
HR Support
- FAQ
- Automating simpler processes like time-off
- Displaying & validating employee information like contact details etc. from HRIS
Compliance
- FAQ
- Guided dialog conversations to help identify the right action to be taken given a situation
Finance
- FAQ
- Requesting for payroll or benefit related information

integrating power virtual agents with enterprise apps

If you have seen or used Microsoft Flow (Power Automate) before, you are in business! That’s because Power Virtual Agents can seamlessly connect with Power Automate flows to integrate with a wide variety of enterprise applications.

Few examples of integrations that you can use today are:

SharePoint, OneDrive, Outlook, etc
Dynamics 365
Salesforce
ServiceNow
Freshdesk
Zendesk
Workday HCM
Box
DocuSign
Jira

And many many more…

making power virtual agents accessible to your employees

At the time of this writing, Power Virtual Agents can be deployed on a variety of different messaging applications.

What’s exciting us the most is the ability to launch an agent on Microsoft Teams on the click of a button!

Once you build your agent, you can enable it on Teams and get your pilot group to start using and testing the bot. With any chatbot, the crucial step is to collect lots of feedback from your pilot or early users before you go to a larger audience.

securing your power virtual agents

This part is a bit technical as it does involve some knowledge of Active Directory. However, once you have the configuration in place, you can protect your sensitive data and transactions by ensuring that an employee is properly signed in before the dialogue is executed.

conclusion

With Power Virtual Agents, any enterprise with Office 365 can start building chatbots to automate a wide variety of use cases. Something that was once considered as the complex is now democratized – true to Microsoft’s mission of empowering every person and every organization to achieve more.

With this level of proliferation of chatbots within an enterprise, we see that the transition from apps to conversational bots is all set to increase in pace. Many enterprises are using Acuvate’s BotCore for a similar end.

BotCore is a chatbot middleware and aggregation platform that helps customers to build virtual agents and, as well, connect existing bots within a single enterprise persona. Together, Power Virtual Agents and BotCore can aid in an AI-driven Digital Transformation within the enterprise. At Acuvate we are looking forward to automating and solving complex enterprise problems with our customers using Power Virtual Agents!

If you’d like to learn more about this topic, please feel free to get in touch with one of our experts for a personalized consultation.

The post Why are we excited about Power Virtual Agents? appeared first on BotCore.

Comparing The Top Bot Development Frameworks

Abhishek Shanbhag — Fri, 05 Oct 2018 14:21:00 +0000

Comparing The Top Bot Development Frameworks

Chatbots are noticeably one of the most popular AI technologies. In the past few years, chatbots have been transforming customer and employee experience, simplifying business workflows and reducing costs. With the rise in demand for chatbots, several frameworks and chatbot platforms have influxed the market. Enterprise leaders who participate in our Build-A-Bot workshops, often seek help in understanding the functionalities of different bot frameworks and platforms.

This blog differentiates between a chatbot development framework and chatbot platform, enlists some of the major bot development frameworks and their key features.

how is it different from a bot development platform?

Most people confuse bot framework with a bot platform or use the word interchangeably

A bot framework is a which helps you develop chatbots with predefined functions and classes. Frameworks are usually used by developers as they involve some programming or coding. They provide some predefined set of tools for faster development of bots.

An enterprise-grade chatbot platform helps you build, train and manage chatbots. It allows non-technical users to build bots without any coding or programming knowledge.

A platform is where the bot is deployed, run and made to perform actions as requested by users. Whereas the framework helps develop and keep together all constituents of a bot. It involves predefined functions and tools that expedite code writing and bot deployment.

major bot development frameworks

Now that we have differentiated between bot development frameworks and platforms, let’s deep dive into some of the most popular bot frameworks and their different features and capabilities.

Microsoft Bot Framework

Microsoft Bot Framework is one of the most comprehensive frameworks for building enterprise chatbots. You can build a simple Q&A bot or a sophisticated virtual assistant.

It is not only intelligent and feature-rich, but it’s also flexible and scalable. Developers can build bots that interact with users in a natural language. This is enabled by Microsoft’s Language Understanding Intelligent Service (LUIS) which extracts intents and entities from conversations. With LUIS, you can constantly improve the natural language models.

The Bot Connector feature of this framework allows bot integration of a variety of platforms such as Slack, Facebook Messenger, Telegram, Webchat, SMS, email, Skype etc.

It also leverages Microsoft’s QnA Maker which allows you to build basic QnA bots based on existing FAQ URLs, structured documents and product manuals.

One of the main advantages of the framework is that it supports Azure Bot Service. Azure allows you to quickly respond to user queries, even if there is high volume. And by using Azure Bot Service, you only have to pay for messages delivered using the Premium channel. In addition, with the service you can have complete ownership and control over data.

Another benefit of this framework is that it provides an open source SDK to build and test chatbots. You can also test and debug bots with Microsoft’s desktop application – Bot Framework Emulator.

Learn More: Why CIOs should consider Microsoft Bot Framework to build Enterprise Bots

WIT.AI

Facebook’s Wit.ai is a natural language bot development framework which enables developers to build both voice and text based bots on virtually any messaging platform of their choice.

It also allows developers to build voice interface for their apps. Moreover, the platform also shares the bot learnings with the developer community who can leverage it to further enhance the user experience.

Additionally, Wit.ai is an effective solution for home automation. It can control any smart device including home appliances and wearables.

Pre-build entities like temperature, URLs, emails, etc. make Wit.ai an excellent virtual assistant. However, the developers may have to work on refining engine training which currently takes a little long.

Dialogflow

Dialogflow, previously known as API.AI, runs on the Google Cloud platform. Powered by Google’s machine learning, it enables bot to understand the intent of the user and respond in the most accurate manner. It also takes user-machine interaction to a new level with voice and text-based conversational interfaces.

It can be integrated with any platform including Google Assistant, Alexa, Cortana, Facebook Messenger, Slack, websites and many more. Dialogflow also supports almost all types of devices such as wearables, phones, car audio, smart devices etc. This means you can connect with your users irrespective of which devices or platforms they’re using. With its ability to support over 20 languages, it helps you expand your global reach.

Another benefit that this framework offers is that it allows fast coding, thus allowing quicker time-to-market.

Botpress

Botpress is an open-source bot development framework built for the developers’ community. The framework is 100% based on Javascript. Since it’s based on a modular architecture, it’s easy to continuously add new features to it.

Botpress is quite flexible in terms of hosting. Depending on their business requirements, users can host it on their enterprise systems, on-premise or on the cloud environment. Also, it’s one of the most user-friendly frameworks. It doesn’t require a user to have the technical knowledge to manage it after it’s deployed.

It allows required customization and facilitates limitless and easy integration with third-party applications and APIs. This also means that users can interact with Botpress bots on all major messaging platforms.

In addition, it allows you to monitor bot application and performance. It records user-bot interaction and with analytics, it allows you to make the required changes to make your bots more intelligent.

RASA Stack

Rasa is another open-source framework which is powered by machine learning. It can be customized fully which makes it a fit choice in enterprise architecture. There are two main components of this framework – Rasa NLU and Rasa Core. Rasa NLU is a natural language processing tool which classifies intents and extracts entities in chatbots. It analyses free text and takes out structured data from it. For example – address, date, numbers etc. Rasa Core uses intents and entities of Rasa NLU to create a reply dialogue. The deep learning technology empowers it to conduct complex conversations.

Rasa’s powerful and intuitive interface facilitates faster training and improves user experience.

Conclusion

These are some of the leading bot development frameworks available today. Every framework has its own pros and cons. It would be difficult, or rather unfair to comment on which one is the best. There is no “best” bot framework in its absolute sense.

Choosing the right bot framework depends on your business needs and technological landscape.

What’s sure is that the conversational bots are here to stay. They’re going to change the face of customer service, employee productivity and business workflows in the coming years.

Acuvate’s own enterprise chatbot builder platform, BotCore can be deployed both on cloud and on-premise environments and helps you deploy enterprise chatbots, train, and administer them according to your needs

If you’re planning to deploy chatbots for your business and need guidance in choosing a powerful bot framework, feel free to get in touch with one of our chatbot experts for a quick consultation.

FREE EBOOK

a guide to choosing an enterprise bot builder platform

The post Comparing The Top Bot Development Frameworks appeared first on BotCore.