IBM, University of Michigan Partner on Voice-Enabled Academic Adviser

IBM and the University of Michigan are collaborating on a speech-enabled system that will provide academic advice for undergraduate computer science and engineering students at the Ann Arbor, Mich.-based university.

Dubbed Project Sapphire, the system will use conversational speech to provide academic scheduling, course, and career guidance and similar advice, much like a human academic adviser, according to Satinder Singh Baveja, a professor of computer science and engineering and director of the university's artificial intelligence lab.

Though there are several other speech-enabled systems, like Apple's Siri and Microsoft's Cortana, the interaction of these systems is not like human interaction, according to Baveja. "Lots of progress is being made, but we're still very far away from building natural language AI; it's a very high bar. We're going to try to build a conversational system that will start with curricular choice."

The Sapphire development team will capture and annotate large volumes of approved recorded human-to-human conversations between undergraduates and their advisers on topics such as course selection, career advice, extracurricular activities, and homework resources. The team will then use these conversations to train the system to respond to interactions with students and ultimately learn how to automatically navigate and successfully reply in conversations with those using the system.

Project Sapphire will apply probabilistic and statistical methods of reasoning that understand conditions and context. By employing deep learning, machine learning, reinforcement learning, natural language understanding, knowledge representation, emotion analysis, and software technologies, the cognitive system will be trained using the recorded human-to-human conversations and will continue to learn and improve with increased interaction.

"Human-to-machine interactions, similar to human-to-human conversations, are rarely confined to one question and one answer," said David Nahamoo, IBM fellow and chief technologist for conversational systems, IBM Watson, in a prepared statement. "They involve multiple turns of a conversation with responses that can be imprecise and unclear, making it difficult to simulate the human experience."

When the project is completed, students would be able to ask Sapphire about courses to take for certain degrees, as well as general questions on career choices, clubs, activities, and other related subjects.

The system will be designed to be very task-oriented to evaluate the student's performance and compare it to the performance of similar students with similar courses or on similar academic paths to provide the best advice, Baveja adds.

Baveja stresses that Sapphire is being designed to augment human advisers, not to replace them. However, he added that human advisers are constrained by time, producing a bottleneck in the current academic system.

Additionally, Sapphire will have access to 20 years of historical students' performances and be able to process it at a much faster speed than a human adviser ever could.

The system would record conversations and at any point, it could hand the session over to a human. One of the researchers involved, Emily Mower Provost, an assistant professor of computer science and engineering, studies emotional cues. She'll work on enabling Project Sapphire to recognize when students need a human adviser, even if they're not asking for one.

The automated academic adviser is one application of what the team envisions as a platform technology. Project Sapphire's resulting innovations could be embedded into cognitive systems across many industries to improve how they learn and codify human expertise, understand user intent and context, and deliver appropriate responses that direct conversations towards stated goals.

Project Sapphire is expected to take a couple of years to complete.

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues
Related Articles

IBM and Hilton Field Test a Robot Concierge

Connie is Hilton's robot concierge that incorporates IBM's Watson cognitive computing and speech APIs to help hotel guests.

IBM and SoftBank Introduce Japanese Watson APIs

Watson's capabilities can now be used to develop cognitive computing and speech services in Japan.