Research Group Projects and Descriptions

The Speech + Mobility group uses speech technologies and portable devices to enhance human communication and make digitized audio more useful as a data type. Our focus is on developing novel applications, user interfaces, and services to exploit computer speech processing for interacting with and through computers far removed from keyboards and monitors.

Active Messenger filters email messages according to dynamic rules, and automatically routes messages to a variety of wired and wireless delivery channels. It observes traffic from a user by knowing which devices have been used to originate or to respond to messages, recent log-ins, and caller ID when checking voice or email over the phone. Its goal is to make sure that desired messages always reach the subscriber, while paring messages down when the user is less reachable. Active Messenger also acts as a proxy, hiding the identity of the multiple device addresses at which the subscriber may be found. More than just a router, Active Messenger is a dynamic process which monitors a message's progress through various channels over time.

Alumni Contributor(s): Stefan Marti

We often have 'idle' time during which we consume various types of audio media (radio, audio books, MP3s, CDs, podcasts)—while driving to work, or waiting at a cafe for a friend. Are We There Yet? helps an audio-playing device put together a listening program that will neatly fit within, for example, your bus ride home. It has audio time-compression techniques to match the playback duration to the user's available listening time.

The Beat Browser is a music and sound browser designed to give users fast feedback while browsing arbitrarily large collections of audio. The goal of Beat Browser is to give the user a sense of exploring “live” and continuous audio, while rapidly moving between sources by mouse. The primary design goal is to facilitate rapid browsing of music by genre, artist, and album in order to interactively build playlists. Beat Browser is a product of larger design strategies identified to build efficient browsing engines, applicable beyond sound.

Why do our advanced cell phones still ring at completely inappropriate times, such as at the theatre, at the movies, or during family dinners? How would a person contact me in an emergency if I have turned off my cell phone because I am in a meeting? In this work, we explore ways to make mobile communication devices socially intelligent, both in their internal reasoning and in how they interact with people. We propose the concept of an Autonomous Interactive Intermediary that helps the user manage her mobile communication devices in a socially appropriate way.

Alumni Contributor(s): Rachel Kern, Stefan Marti and Matt Hofmann

The Democratic Public Address system is a distributed mesh network of decidedly inexpensive speaker nodes that play audio messages. Each node contains a microphone and speaker, and can be used to program the message that will be rebroadcast by the network. With the DPA, it is possible to easily and quickly deploy a freely accessible messaging platform in public spaces. Nodes are cheap enough that the deployments can be considered disposable, and usable in spaces where authoritarian presences restrict democratic speech.

Globetoddler is a new project that focuses on remote awareness and shared experiences from children's point of view. The project aims to develop a technology and software solution that makes it easier, as well as more fun and rewarding, for young children (3-6 years) to interact with their remote, traveling parents. Newly conducted interviews indicate a need for a system with very different interfaces and interaction styles on the the two sides: on the parent's side the system consists of a fairly simple mobile application, whereas the child's side consists of a combination of a physically interactive toy interface and a guiding virtual avatar.

Going My Way is a work in progress consisting of an application that learns our everyday routes and provides directions from points along those paths. The phone-client application, based on our custom-built Contella1 platform, logs GPS information and translates this into a route model. It sends this route information to our custom server application, integrated with Microsoft’s MapPoint, and computes the shortest path between your everyday route and your desired destination. Going My Way then provides directions, if available, based on personal landmarks rather than street names and intersections. We aim to reduce cognitive load by simplifying directions: guiding you to your destination with the knowledge of where you’ve been.

Alumni Contributor(s): Chaochi Chang

Monkey Business is a system to keep distributed members of a group aware of each other’s presence and activities in a light-hearted manner, while striving to remain non-intrusive. The system also aims to facilitate unplanned and informal communication among distributed colleagues. It consists of a network of animatronic agents, specifically monkeys, which are situated in the offices or rooms of each member of a group. Through subtle movements, gestures, and sounds, the monkeys indicate the current activities of the other members of the group. The monkeys are meant to be ambient, at the periphery of one’s attention; but they can also be used more proactively as communication mechanisms, and promote informal exchanges among members of a distributed team.

Alumni Contributor(s): Rachel Kern and Stefan Marti

MonkeyToGo is a mobile extension of Monkey Business, our animatronic, remote- awareness network. Cell phones running our MonkeyToGo application can easily be added to the network in order to interact with and receive information from the agents in real time. If active, the connected cell phone vibrates and plays short animations and sound snippets in order to inform its user of ongoing activities at the remote locations. The MonkeyToGo user can also use the cell phone to "tickle" (call for attention) or to set up a voice connection with other monkey agents. Several cell phones can be added at the same time.

People's daily lives are impacted by a great deal of dynamic environmental information, such as weather and traffic. Though most of this information is now available on the Internet, there isn't an easy way to access it while mobile. In addition, people do not always need to know this kind of information unless there is a significant change that may impact their current or future activities. Ringing in the Rain proposes a distributed, multi-agent architecture that uses GPS-enabled cell phones to build a mobile service development framework. The goal of this framework is to build mobile services to deliver timely changes in environmental information that could impact a user's current or future activities. A weather warning system is developed based on this framework as a demonstration to inform people to leave or detour in advance to avoid being caught in the rain.

We are developing a tool to allow users to place and recall virtual sound objects (music files, phone numbers) at 3-D spatial coordinates around us. As these locations are in fact imaginary, it frees us to explore visually independent interaction through audio. Two turns of the wrist, for pitch and yaw orientation, and a scroll of the thumb to provide a distance value, and we are exploring different notions of “file paths” and “memory addressing.”

To maintain our relationships with family and friends, it’s important to share our daily lives. TimelessWords is an integrated, calendar-based life-sharing system to enhance emotion-driven daily communication. Users can easily pre-record voice messages via a handheld device, and send the messages to other community members. TimelessWords users have a shared calendar. Messages will be delivered at scheduled moments to express timeless care.

Our project, ToDoGo, learns by observing patterns of the user's location to answer the questions "What should I do?", "Where am I going", and "How do I get there?". For example, if you need to send a letter, you'll typically do this at a post office or drop box. ToDoGo uses common-sense knowledge and data about local points of interests to associate everyday activities with locations in the Boston area. From this, ToDoGo provides a wide range of applications, including a just-in-time reminder, activity/destination/path recommendations, path optimization, and distributed household management.

Communication is much more than the direct transfer of information. It is an interactive collaborative act including potentially rich verbal and non-verbal cues. To date, telecommunication has focused on verbal communication at a distance, restricting the non-verbal situational information. With the existing infrastructure, it is necessary to first contact a person before receiving feedback on what they are doing. Current communication devices do not provide a persistent sense of presence of those with whom we connect, nor do they create opportunities for additional communication with them. WatchMe—a platform for mobile communication and awareness in the form of a watch—addresses these limitations.

Alumni Contributor(s): Natalia Marmasse

This project describes the design and development of a system that will help to increase safety and security using cell phones. This system has two main components: a master phone application to assist people who need to take care of their loved ones, and a slave phone application to provide help to care-recipients who need attention from their caregivers. This system applies location awareness (GPS), awareness of social activities (communication activity and proximity with close peers), and peer-to-peer data communication as its core technologies.