BIRLA INSTITUTE OF TECHNOLOGY & SCIENCE, PILANI

WORK INTEGRATED LEARNING PROGRAMMES

COURSE HANDOUT

Part A: Content Design

Course Title Design of Conversational Experiences

Course No(s) SE ZG530
Credit Units 4
Course Author Dr. Shreyas Rao
Version No 1.0
Date June 2024

Course Description
Cognitive virtual assistant (CVA): Use-cases; Classification of conversational AI platforms;
Architecture of Conversational Platform; Deployment and Pricing models; Platform landscape;
Designing Bots: Bot Architecture; Bot Anatomy; Design process overview; Branding, Personality,
and Human Involvement; Conversation; Rich interactions; Engagement methods; Use case
definition and exploration; Conversation scripting; Context and Memory; User testing; Designing
Voice User Interfaces(VUI): Conversational Voice User Interface(VUI); VUI Designer; VUI design
principles; Designing effective process and dialogue; Personas, Avatars, Actors; Speech recognition
technology; Advanced VUI Design; User testing; Development: Building and deploying
conversational AI assistants (voice assistants & chatbots) using cloud native / open source
platforms such as Google Dialogflow, RASA or MS Bot framework; Bot Discovery and installation;
Monetization; Analytics and Continuous improvement; Trends: SuperBot Platforms; Multiplatform
Bots; Identity consolidation; Voice-enabled Devices – Smart Homes and Smart Cars as example
environments.

Course Objectives
No Course Objective

CO1 Introduce the foundations of conversational AI and platforms

CO2 Understand key design concepts for Bots and Voice User Interfaces

CO3 Design Conversational Experiences across different use cases

 CO4 Build AI chatbots and voice bots using practical toolkits

Text Book(s)
T1 “Designing Bots – Creating Conversational Experiences” by Amir Shevat. Publisher:
O’Reilley, 2017
T2 “Designing Voice User Interfaces – Principles of Conversational Experiences” by Cathy
Pearl. Publisher: O’Reilley, 2017

Reference Book(s) & other resources

“The Definitive Guide to Conversational AI with Dialogflow and Google Cloud” by Lee
R1
Boonstra. Publisher Apress, 2021

R2 “Cognitive Virtual Assistants Using Google Dialogflow” by Navin Sabharwal, Amit

Agrawal. Publisher: Apress, 2020
“Conversational AI - Chatbots that work” by Andrew Freed. Publisher: Manning, 2021
R3

R4 “Build Better Chatbots - A Complete Guide to Getting Started with Chatbots” by Rashid
Khan, Anik Das. Publisher: Apress, 2018

WR1 Enterprise Conversational AI Platform -

https://www.cxtoday.com/data-analytics/gartner-magic-quadrant-for-enterprise-
conversational-ai-platforms-2022/

WR2 Amazon Lex - https://docs.aws.amazon.com/pdfs/lexv2/latest/dg/lex2.0.pdf

WR3 Google DialogFlow

https://landbot.io/blog/chatbot-using-dialogflow-integration
https://medium.com/@dipan.saha/chatbot-development-made-easy-creating-a-
simple-bot-with-dialogflow-ade69caac37d

WR4 Multi-lingual Bots

Amazon Lex - https://aws.amazon.com/blogs/machine-learning/building-a-multilingual-
question-and-answer-bot-with-amazon-lex/

WR5 LLMs in Conversational AI - https://www.analyticsvidhya.com/blog/2023/07/llms-in-

conversational-ai/
No code LLM-based Chatbot - https://flowiseai.com/
Modular Content Structure

M1 - Introduction (4 hours)
 Conversational Interfaces
 Conversational Platforms: Characteristics, Classification and Landscape
 Virtual Assistant vs Chatbots
 Cognitive Virtual Assistants (CVA)
 Use cases for virtual assistants, chatbots and CVA
 Architecture of Conversational Platforms
 Overview of Deployment and Pricing models
 Enterprise Conversational AI platforms
o Natural-language-portfolio centric

o Business automation centric

o User-experience centric
M2 – Bot Basics (2 hours)
 Types of Bots
 Bot platforms
 Bot Anatomy
 Branding, Personality, and Human Involvement

M3 – Designing bot conversations (4 hours)

 Onboarding
 Functionality Scripting
 Task-led and topic-led conversations
o Divergent flows and course correction
o Entity extraction
o Intent mapping
o Conversational controls
o Stories/ Flows
 Decoration
 Rich Interactions
 Context and Memory
 Bot Discovery and Installation
 Engagement Methods

M4 – Case studies on bot design (2 hours)

 Case Studies of PTOBot and VacationBot
 Use Case Definition and Exploration
 Conversation Scripting
 Designing and Testing

M5 – Bot Building and Deployment (4 hours)

 Bot building overview
 Comparison of bot builder frameworks and tools
o Open source vs commercial
o No code vs low code
 Bot deployment across multiple channels
  Demo of end-to-end conversation management using bot builders
 Monetization
 Analytics and Continuous Improvement

M6 – Designing Voice User Interfaces (4 hours)

 Conversational Voice User Interface (VUI)
 VUI Designer
 VUI Design Principles
 Conversational Design
o Setting user expectations
o Design tools
o Confirmations
 Personas
 Avatars
 Pros and cons of Avatars
 Actors
 Speech Recognition Technology

M7 – Advanced VUI design (2 hours)

 Branching
 Disambiguation

 Negation Handling

 Sentiment Analysis and Emotion Detection

 Text-to-Speech Versus Recorded Speech

 Advanced Multimodal and NLU

M8 – Building, Testing and Deploying VUI Apps (4 hours)
 Overview of building voice assistants using cloud native / open source platforms
 Building a VUI Application
o Utterance
o Intents
o Entities
o Fulfilment
o Integrations
o Training
o Validation
 Monetization of VUI application
 Analytics in VUI application
 User testing
 Early-Stage Testing

 Usability Testing

 Performance Measures
 Deployment channels
M9 – Building Multi-Lingual Virtual Agents (2 hours)
 Agent languages
 Building a Multi-lingual virtual agent
 Working with translation service
 Multi-language training and testing
 Discussion on frameworks and tools that support multi-lingual agent development

M10 – Trends (4 hours)

 Super Bot Platforms
 Multiplatform Bots
 User Identity Consolidation
 Voice-enabled Devices: Smart Homes and Smart Cars as example environments
 Building chatbots using Large Language Models (LLMs)
 RAG model for chatbot development
 Multimodality for improved Customer Experience (CX)

Learning Outcomes: At the end of the course, the student will be able to

No Learning Outcomes

LO1 Obtain understanding of the applications of Cognitive virtual assistants and chatbots

LO2 Apply knowledge of the design and process flows necessary for creating conversational
experiences

LO3 Develop hands-on experience in implementation and deployment of chatbots and voice
assistants

LO4 Gather knowledge of practical ways of testing, releasing, analysing and monetizing the
conversational applications

Part B: Contact Session Plan

Academic Term First Semester 2024-2025

Course Title Design of Conversational Experiences
Course No SE ZG530
Lead Instructor Dr. Shreyas Rao
Course Contents

Contact List of Topic Title Text/Ref Book/external

Session (from content structure in Part A) resource

1 Introduction
 Conversational Interfaces R2 – Chapter 1
 Conversational Platforms: Characteristics, Classification T1 – Chapter 4
and Landscape Lecture Notes
 Virtual Assistant vs Chatbots
 Cognitive Virtual Assistants (CVA)
 Use cases for virtual assistants, chatbots and CVA

2 Introduction (contd…)
 Architecture of Conversational Platforms R2 – Chapter 1
 Overview of Deployment and Pricing models R3 – Chapter 1
 Enterprise Conversational AI platforms WR1
Lecture Notes
o Natural-language-portfolio centric

o Business automation centric

o User-experience centric

3 Bot Basics
 Types of Bots T1 - Chapters 2,3,5, 6
 Bot platforms
 Bot Anatomy
 Branding, Personality, and Human Involvement

4 Designing Bot Conversations

 Onboarding T1 – Chapter 8
 Functionality Scripting R4 - Chapters 3, 4
 Task-led and topic-led conversations
o Divergent flows and course correction
o Entity extraction
o Intent mapping
o Conversational controls
o Stories/ Flows
 Decoration

5 Designing Bot Conversations (contd…) T1 – Chapters 9, 10, 11,12

 Rich Interactions
 Context and Memory
 Bot Discovery and Installation
 Engagement Methods
 6 Case Studies on Bot design T1- Chapters 14, 15, 16,
 Case Studies of PTOBot and VacationBot 17
 Use Case Definition and Exploration
 Conversation Scripting
 Designing and Testing

7&8 Bot Building and Deployment T1 - Chapters 13, 18, 19

 Bot building overview R4 - Chapters 3,4
 Comparison of bot builder frameworks and tools Lecture Notes
o Open source vs commercial
o No code vs low code
 Bot deployment across multiple channels
 Demo of end-to-end conversation management using
bot builders
 Monetization
 Analytics and Continuous Improvement

9 Designing Voice User Interfaces (VUI) T2 – Chapters 1, 2

 Conversational Voice User Interface (VUI)
 VUI Designer
 VUI Design Principles
 Conversational Design
o Setting user expectations
o Design tools
o Confirmations

10 Designing Voice User Interfaces (contd…) T2- Chapters 3, 4

 Personas
 Avatars
 Pros and cons of Avatars
 Actors
 Speech Recognition Technology

11 Advanced VUI Design T2 - Chapter 5

 Branching
 Disambiguation

 Negation Handling

 Sentiment Analysis and Emotion Detection

 Text-to-Speech Versus Recorded Speech

 Advanced Multimodal and NLU

12 Building VUI applications R3 - Chapter 2

 Overview of building voice assistants using cloud WR2
native / open source platforms WR3
 Building a VUI Application Lecture Notes
o Utterance
 o Intents
o Entities
o Fulfilment
o Integrations
o Training
o Validation
 Monetization of VUI application
 Analytics in VUI application

13 Testing and Deploying VUI applications T2 – Chapter 6

 User testing Lecture Notes
 Early-Stage Testing

 Usability Testing

 Performance Measures

 Deployment channels

14 Building Multi-Lingual Virtual Agents R1 – Chapter 8

 Agent languages R2 – Chapter 3
 Building a Multi-lingual virtual agent WR4
 Working with translation service Lecture Notes
 Multi-language training and testing
 Discussion on frameworks and tools that support multi-
lingual agent development

15 Trends - I T2 – Chapter 6
 Super Bot Platforms Lecture Notes
 Multiplatform Bots
 User Identity Consolidation
 Voice-enabled Devices: Smart Homes and Smart Cars as
example environments

16 Trends – II WR5
 Building chatbots using Large Language Models (LLMs) Lecture Notes
 RAG model for chatbot development
 Multimodality for improved Customer Experience (CX)

Experiential Learning

Lab. Module Session Experiment Tools list (Indicative only)

No No. No.
1 5 7 or 8 Demonstrate the design and Free tools - Wit.ai (from
development of a text-based bot for a Meta), Botkit
simple use case. Deploy the bot on
any one channel.
2 8 12 Demonstrate the design and Amazon Lex (works with
development of a virtual agent AWS Free tier), Google
answering queries in both text and DialogFlow
 voice modes (English language).
Deploy on any two channels.
3 9 14 Extend the second experiment to Engati (supports 30+ Indian
include one more language languages), Amazon Lex
(preferably Indian origin). Deploy on (Hindi), Google Dialogflow
any two channels. (Hindi).
4 10 16 [Optional] https://flowiseai.com/
Create a simple chatbot using any [Free]
large language model, employing no-
code or low-code methods.

Evaluation Scheme

Evaluation Name Type Weight Duration Day, Date, Session,

Component (Quiz, Lab, Project, (Open book, Time
Midterm exam, End Closed book,
semester exam, etc) Online, etc.)

EC – 1 Assignment I 10%

Assignment II 15%

Assignment III 15%

EC – 2 Mid-term Exam Closed book 30% 2 hours

EC – 3 End Semester Exam Open book 30% 2 hours

Note - Evaluation components can be tailored depending on the proposed model.

Assignments

Assignment I (10%)
Task: Comparative analysis of Bot Builders.
Sub-tasks:
1. Select one open source bot builder (Ex: Wit.ai) and one commercial bot builder (Ex: Amazon Lex).
2. Write a detailed report comparing the two selected bot builders, based on the following criteria:
a. Ease of Development (No code, drag and drop functionality etc.)
b. Features and capabilities.
c. Integration options with other tools and platforms.
d. Community support and documentation.
e. Real world use cases for the bot builder and its usage in the Industry.

Assignment II (15%)
Task: Design, build, and deploy a text-based chatbot using any open source tools.
Sub-tasks:
1. Design the bot persona which includes name, environment, audience, task at hand, runtime
variations, service branding, values, derived personality etc. [Refer T1 – Chapter 15]
2. Design the conversation flow, which includes identifying the utterances, intents, entities,
conversation flows (main flow and alternate flows), integration aspects etc.
3. Build and test the bot.
4. Deploy the bot in any one channel (Web / Facebook etc.).
Note: The students are encouraged to develop the bot on their own, and not use any pre-built
templates that are available as part of the bot builder.

Assignment III (15%)

Task: Design, build, and deploy a multi-lingual voice-based AI assistant, using any tool of your choice.
Sub-tasks:
1. Identify the languages that the assistant should support (English plus another language).
2. Design the conversation flows, which includes identifying the utterances, intents, entities,
fulfilment mechanisms, and integrations.
3. Build the assistant.
4. Test the assistant for multi-language support.
5. Deploy the assistant in any one channel (Web / Facebook etc.).
Note: The students are encouraged to develop the assistant on their own, and not to use any pre-
built templates that are available (Ex: Google Dialogflow and Amazon Lex provides templates for
common use cases such as airline services, doctor’s appointment, insurance, e-commerce etc.)

Important Information
Syllabus for Mid-Semester Test (Closed Book): SL No. 1 to 8
Syllabus for Comprehensive Exam (Open Book): All topics given in plan of study
Evaluation Guidelines:
1. EC-1 consists of one quiz and three assignments. Announcements regarding the same will be
made in a timely manner.
2. For Closed Book tests: No books or reference material of any kind will be permitted.
Laptops/Mobiles of any kind are not allowed. Exchange of any material is not allowed.
3. For Open Book exams: Use of prescribed and reference text books, in original (not
photocopies) is permitted. Class notes/slides as reference material in filed or bound form is
permitted. However, loose sheets of paper will not be allowed. Laptops/Mobiles of any kind
are not allowed. Exchange of any material is not allowed.
4. If a student is unable to appear for the Regular Test/Exam due to genuine exigencies, the
student should follow the procedure to apply for the Make-Up Test/Exam. The genuineness
of the reason for absence in the Regular Exam shall be assessed prior to giving permission to
appear for the Make-up Exam. Make-Up Test/Exam will be conducted only at selected exam
centers on the dates to be announced later.

It shall be the responsibility of the individual student to be regular in maintaining the self-study
schedule as given in the course handout, attend the lectures, and take all the prescribed evaluation
components such as Assignment/Quiz, Mid-Semester Test and Comprehensive Exam according to
the evaluation scheme provided in the handout.