Move Over, Siri! University of Southern California Researchers Develop Improv-Based Chatbot

What would conversations with Alexa be like if she was a regular at The Second City?

Jonathan May, research lead at the USC Information Sciences Institute (ISI) and research assistant professor of computer science at USC's Viterbi School of Engineering, is exploring this question with Justin Cho, an ISI programmer analyst and prospective USC Viterbi Ph.D. student, through their Selected Pairs Of Learnable ImprovisatioN (SPOLIN) project. Their research incorporates improv dialogues into chatbots to produce more engaging interactions.

The SPOLIN research collection is made up of over 68,000 English dialogue pairs, or conversational dialogues of a prompt and subsequent response. These pairs model yes-and dialogues, a foundational principle in improvisation that encourages more grounded and relatable conversations. After gathering the data, Cho and May built SpolinBot, an improv agent programmed with the first yes-and research collection large enough to train a chatbot.

The project research paper, "Grounding Conversations with Improvised Dialogues," was presented on July 6 at the Association of Computational Linguistics conference, held July 5-10.

Finding Common Ground

May was looking for new research ideas in his work. His love for language analysis had led him to work on Natural Language Processing (NLP) projects, and he began searching for more interesting forms of data he could work with.

"I'd done some improv in college and pined for those days," he said. "Then a friend who was in my college improv troupe suggested that it would be handy to have a 'yes-and' bot to practice with, and that gave me the inspiration--it wouldn't just be fun to make a bot that can improvise, it would be practical!"

The deeper May explored this idea, the more valid he found it to be. Yes-and is a pillar of improvisation that prompts a participant to accept the reality that another participant says ("yes") and then build on that reality by providing additional information ("and"). This technique is key in establishing a common ground in interaction. As May put it, "Yes-and is the improv community's way of saying 'grounding.'"

Yes-ands are important because they help participants build a reality together. In movie scripts, for example, maybe 10-11% of the lines can be considered yes-ands, whereas in improv, at least 25% of the lines are yes-ands. This is because, unlike movies, which have settings and characters that are already established for audiences, improvisers act without scene, props, or any objective reality.

"Because improv scenes are built from almost no established reality, dialogue taking place in improv actively tries to reach mutual assumptions and understanding," said Cho. "This makes dialogue in improv more interesting than most ordinary dialogue, which usually takes place with many assumptions already in place (from common sense, visual signals, etc.)."

But finding a source to extract improv dialogue from was a challenge. Initially, May and Cho examined typical dialogue sets such as movie scripts and subtitle collections, but those sources didn't contain enough yes-ands to mine. Moreover, it can be difficult to find recorded, let alone transcribed, improv.

The Friendly Neighborhood Improv Bot

Before visiting USC as an exchange student in Fall 2018, Cho reached out to May, inquiring about NLP research projects that he could participate in. Once Cho came to USC, he learned about the improv project that May had in mind.

"I was interested in how it touched on a niche that I wasn't familiar with, and I was especially intrigued that there was little to no prior work in this area," Cho said. "I was hooked when Jon said that our project will be answering a question that hasn't even been asked yet: the question of how modeling grounding in improv through the yes-and act can contribute to improving dialogue systems."

Cho investigated multiple approaches to gathering improv data. He finally came across Spontaneanation, an improv podcast hosted by prolific actor and comedian Paul F. Tompkins that ran from 2015 to 2019.

With its open-topic episodes, about a good 30 minutes of continuous improvisation, high quality recordings, and substantial size, Spontaneanation was the perfect source to mine yes-ands from for the project. The duo fed their Spontaneanation data into a program, and SpolinBot was born.

"One of the cool parts of the project is that we figured out a way to just use improv," May explained. "Spontaneanation was a great resource for us, but is fairly small as data sets go; we only got about 10,000 yes-ands from it. But we used those yes-ands to build a classifier (program) that can look at new lines of dialogue and determine whether they're yes-ands."

Working with improv dialogues first helped the researchers find yes-ands from other sources as well, as most of the SPOLIN data comes from movie scripts and subtitles. "Ultimately, the SPOLIN corpus contains more than five times as many yes-ands from non-improv sources than from improv, but we only were able to get those yes-ands by starting with improv," May said.

SpolinBot has a few controls that can refine its responses, taking them from safe and boring to funny and wacky, and also generates five response options that users can choose from to continue the conversation.

SpolinBot #Goals

The duo has a lot of plans for SpolinBot, along with extending its conversational abilities beyond yes-ands. "We want to explore other factors that make improv interesting, such as character-building, scene-building, 'if this (usually an interesting anomaly) is true, what else is also true?,' and call-backs (referring to objects/events mentioned in previous dialogue turns)," Cho said. "We have a long way to go, and that makes me more excited for what I can explore throughout my PhD and beyond."

May echoed Cho's sentiments. "Ultimately, we want to build a good conversational partner and a good creative partner," he said, noting that even in improv, yes-ands only mark the beginning of a conversation. "Today's bots, SpolinBot included, aren't great at keeping the thread of the conversation going. There should be a sense that both participants aren't just establishing a reality, but are also experiencing that reality together."

That latter point is key, because, as May explained, a good partner should be an equal, not subservient in the way that Alexa and Siri are. "I'd like my partner to be making decisions and brainstorming along with me," he said. "We should ultimately be able to reap the benefits of teamwork and cooperation that humans have long benefited from by working together. And the virtual partner has the added benefit of being much better and faster at math than me, and not actually needing to eat!"

About the USC Viterbi School of Engineering

Engineering began at the University of Southern California in 1905. Nearly a century later, in 2004, the school was named the USC Viterbi School of Engineering, honoring alumnus Andrew J. Viterbi, inventor of the Viterbi algorithm, a key to cell phone technology and numerous data applications. The school is consistently ranked among the top engineering programs in the world. It enrolls more than 7,000 undergraduate and graduate students, taught by 188 tenured and tenure-track faculty, with more than 90 endowed chairs and professorships. The school offers comprehensive programs in all major areas of engineering, from aerospace and astronautical engineering, to biomedical, chemical, civil and environmental engineering, to computer science and materials science, and to electrical engineering, industrial and systems engineering and mechanical engineering. With annual research expenditures exceeding $207M, the school is home to the Information Science Institute and more than 45 other research centers. Its more than 50,000 alumni span the globe. USC Viterbi is also home to the pioneering distance learning program DEN@Viterbi, ranked as the top such program in the nation the last two years. USC Viterbi is the only engineering school in the nation with three top-10 USNWR rankings: Best Graduate Schools, Best Online Graduate Engineering Programs, and Best Online Information Technology Programs. 

IBM UC2 helps people stay connected

Today's organizations are largely global in nature with the workforce being mobile and virtual. In the present environment, it is important for employees to stay connected and collaborate, says Chandru K., Business Manager (Converged Communications Services) at IBM India.
In an interview with Manu Sharma of CIOL Bureau, he spoke on the challenges faced by enterprises and the steps a CIO needs to ensure success of converged communication systems.

CIOL: Role of IBM in Converged Communications?
Chandru K: IBM offers a suite of assessment, design, deployment, integration and managed services to deliver converged communications solutions. IBM offers a uniquely flexible and broad portfolio of services, software and financing capabilities that can be easily customized to address specific needs required to support communications for a diverse and global workforce.
These capabilities are backed by the experience and technical know-how required to provide smooth, successful unified communications and collaboration deployments. Working with a broad network of partners and suppliers (device manufacturers, network equipment vendors, telecommunication service providers and independent software vendors), IBM can offer an extensive range of skills and solutions to meet the client's immediate and long-term business needs for converged communication solutions.

CIOL: What are the present challenges in communication that the enterprises are facing and what does Converged Communications help address/solve?
CK: Organizations are largely global in nature with workforce being mobile and virtual. The focus for organizations is to improve their bottom line that means they are focused on employee productivity. It is important for employees to stay connected and collaborate.
Typical challenges are:
* Communication-caused delay and disruption is a pervasive business problem.
* Communications complexity affects long-term productivity, business process reform, and financial performance.
* Decision support outcomes suffer from the inability to access and collaborate effectively with primary players.
* Resources are improperly used or misallocated because of the complexity of communication.
Converged Communications helps address these challenges and streamline communications.

CIOL: What do you accomplish by deploying Converged Communications, in the short and long terms?
CK: Converged Communications with its emerging class of applications and services helps improve communications. It helps businesses in simplifying communications for their mobile and distributed workforce to improve communication flows, access primary decision makers quickly, enhance collaboration, and improve productivity to positively affect their business. They speed access and improve communication, integrate different device modes and communication applications, and dramatically improve collaboration, allowing organizations to streamline business processes, reach the right resource the first time, and enhance profitability.

CIOL: What are the steps a CIO needs to prioritize to ensure success of Converged Communication systems?
CK: Firstly take a step back and do an assessment of the opportunities What do I spend on tools today? What are my opportunities for savings? And then do a business evaluation. Given those opportunities I uncover, how will I leverage those against my business processes?
Then, a very, very important step and that is to do a network assessment. Is my network ready? And if not, take the step, which is to get the network ready for voice deployments. And when we say network-readiness, we're talking about things such as what sort of network latency do I see in my data network? What are the jitter characteristics of my network? How much bandwidth do I have? And very, very importantly, have I implemented quality of service?
Beyond those steps, customers need to begin to prepare themselves to develop the skills necessary and really to begin to bring together skills between their traditional telephony groups and their data networking groups. And, of course, once you've gotten to that point, then it's time to move into the design and implementation phases.

CIOL: What are the differences between Unified Communications and Unified Messaging?
CK: Unified Communications enables people to find peers or decision makers using a single telephone number or Internet address. It integrates e-mail, instant messaging, and calendaring applications with communications devices and applications—telephony—wired and wireless; voice messaging; and audio, video, and Web conferencing.
Unified Communications applications support advanced presence-sensitivity and find-me capability, and media independence. They are easy to use, with a familiar intuitive interface linked to powerful functions. Finally, they provide voice access to applications and data.
While, Unified Messaging is a subset or one of the Unified Communications applications, Unified Messaging is the integration of different messaging (voice, email and fax) into a single or Unified Message store. They can access this through a variety of devices. Since these are integrated it makes it easier for users to access their message store. For e.g., they can access their voice, fax and email messages through a single touch point, their emails.

CIOL: What are the benefits of using IBM UC2?
CK: IBM offers the essential software, services, strategic alliances and hardware elements that companies need to help people stay connected to applications, data and one another anytime, anywhere. So, users can find an expert, reach out within and beyond business boundaries and collaborate more easily. And, as a result, help organizations improve productivity, speed responsiveness and lower operational costs, while creating a better, more attractive work environment that's easier to manage, extend and enrich.
Unified Communications requires an integrated communications strategy and architecture—enabling the secure combination of voice, video, data, and mobility applications within a robust, intelligent network. IBM, with its industry-leading breadth of solutions, its world-class partnerships, and its understanding of how to map technology to address business challenges, is uniquely able to fulfill the promise of Converged Communications.

CIOL: Why do I need technology from Avaya, AVST or Cisco?
CK: Global Technology Services within IBM focuses on providing infrastructure solution for customers. We work with customers to understand their business problems and check on the best technology to fit this. Working with a broad network of partners and suppliers (device manufacturers, network equipment vendors, telecom service providers and independent software vendors), IBM can offer an extensive range of skills and solutions to meet the client's needs.

CIOL: What is IBM's predictions for Unified Communications?
CK: The Virtual Workplace will become the rule! There will be no need to leave the office. Just bring it [Unified Communications] along. Desk phones and desktop computers will gradually disappear, replaced by mobile devices, including laptops that take on traditional office capabilities. Social networking tools and virtual world meeting experiences will simulate the feeling on being there in-person. Work models will be changed by expanded globalization, and green business initiatives that reduce travel and encourage work at home.
* Instant Messaging and other real-time collaboration tools will become the norm, bypassing e-mail. Just as e-mail became a business necessity, a new generation of workers has a new expectation for Instant Messaging (IM) as the preferred method of business interaction. This will fuel more rapid adoption of Unified Communications as traditional IM becomes the core extension point for multi-modal communications.
* Beyond Phone Calls to Collaborative Business Processes. Companies will go beyond the initial capabilities of IM, like click-to-call and online presence, to deep integration with business processes and line-of-business applications, where they can realize the greatest benefit.
* Interoperability and Open Standards will tear down proprietary walls across business and public domains. Corporate demand for interoperability and maturing of industry standards will force unified communications providers to embrace interoperability. Converged, aggregated, and rich presence will allow businesses and individuals to better find and reach the appropriate resources, removing inefficiencies from business processes and daily lives.
* New meeting models will emerge. Hang up on routine, calendared conference calls. The definition of "meetings" will radically transform and become increasingly adhoc and instantaneous based on context and need. 3-D virtual world and gaming technologies will significantly influence online corporate meeting experiences to deliver more life-like experiences demanded by the next generation workers who will operate more efficiently in this familiar environment.

CIOL: Why is IBM holding this roundtable and what will be the focus? What issues will IBM address at the meet?
CK: IBM in association with CIOL is bringing a live session on Converged Communication. The rountable will focus on the subject, "Enhancing effective business communication". IBM will address issues such as how to communicate and collaborate effectively in today's disparate work environment and how to reduce reliance on external voice, video, date service providers.
The forum will provide an opportunity to discuss the communication challenges organization are facing today and understand how Converged Communication solutions help address these. Customers will have an opportunity to talk to experts and also network with industry peers. IBM will also showcase the actual working of Converged Communication solutions at our customer solution center at Delhi.