The help desk at Classics, Inc., handles more than 100 calls a day for users who need to reset their passwords. The administrator, Akela Wolfe, wants to create a bot to assist users with this task so agents can concentrate on resolving more complex issues. She decides to create the bot using Bot Builder.
Bot Builder bots work with any digital channel, such as Live Chat, Apple Messages for Business, and Instagram. You can use Bot Builder bots on voice channels by setting up an integration in CXone Mpower Virtual Agent Hub.
You can also use bots within CXone Mpower as taskbots. Taskbots allow you to automate certain tasks within the system. For example, you create taskbots to work with Task Assist in Copilot for Agents. Task Assist taskbots can update contact information in a database.
Start learning how to use Bot Builder or explore the implementation process.
Conversational AI
Bot Builder uses an approach called Conversational AI to design and train Bot Builder bots as virtual agents to use in CXone Mpower. This approach combines technologies that let computer software:
- Recognize and decipher human language.
- Comprehend what is being said.
- Determine the right response.
- Respond in a way that mimics human conversation.
Some virtual agents are completely scripted. A developer must try to account for every keyword a customer might use in a given scenario. Then the developer has to script a response for the virtual agent for each of these keywords. As a result, this kind of virtual agent can be very time-consuming to set up and maintain.
Bot Builder bots retain context throughout an interaction. They use artificial intelligence (AI) to predict what contacts
The person interacting with an agent, IVR, or bot in your contact center. want based on training data that provides examples of similar conversations. When training data comes from real conversations between agents and contacts, the bot learns to recognize what a contact wants.
Natural Language Understanding
Bot Builder uses Natural Language Understanding (NLU) to understand what contacts The person interacting with an agent, IVR, or bot in your contact center. say and make accurate predictions about what they mean. It's one of the fundamental technologies behind conversational AI. It's part of Natural Language Processing (NLP), a technology that allows computer programs to interpret and understand human language. NLU is a subset of NLP that focuses on understanding the meaning behind what a human says.
NLU relies on configuration and training that teaches the bot to recognize what contacts mean. This starts with defining the tasks you want the bot to help contacts with, also known as intents. This could be tasks such as updating addresses, providing account balances, or resetting passwords. By creating intents and providing real-world conversation examples of each intent, you help your Bot Builder bot learn to associate what the contact says with what they mean.
Akela Wolfe finds examples of real agent conversations about resetting passwords and uses them to train her new Bot Builder bot. She finds examples where the contact expresses their need in different ways. Some of her examples include the contact saying "I need help changing my password," "My account might be hacked, what can I do," and "My password isn't working, how do I change it"
Key Facts about Bot Builder
- Bot Builder bots can be text-based or voice-based.
- Text-based bots communicate via chat. They require digital channels Various voice and digital communication mediums that facilitate customer interactions in a contact center., which are available with CXone Mpower Digital Experience.
- You can use Bot Builder bots built with Bot Builder with any digital Any channel, contact, or skill associated with Digital Experience. channel supported by Digital Experience. This includes live chat Agents and contacts interact on a real-time basis., SMS Messaging, and all social Public social media interactions, such as on Facebook or Instagram. and messaging Direct interactions using social media like WhatsApp or Facebook Messenger channels.
- Voice-based bots use speech-to-text Also called STT, this process converts spoken language to text. services (transcription) to convert the contact's spoken utterances into text that the bot can analyze. They also use text-to-speech Allows users to enter recorded prompts as text and use a computer-generated voice to speak the content. services to synthesize the responses from the bot into speech in an audio file that is played for the contact.
- To use Bot Builder bots with voice channels , you must have Virtual Agent Hub. Bot Builder for voice is available as a natively-supported virtual agent integration. This option also requires custom scripting.
- You must enable the Chatbot permission in the Digital Experience permissions for the role assigned to the employee account of anyone who will use Bot Builder to create or manage bots.