CMU wearable computers for real-time speech translation

Abstract

Carnegie Mellon's Wearable ComputersLaboratory has built four generations of real -time speech translation wearable computers, culminating in the Speech Translator Smart Module. Smart Modules are a family of interoperable modules supporting real-time speech recognition, language translation, and speech synthesis. In this paper, we examine the effect of various design factors on performance with emphasis on modularity and scalability. A system-level approach to power / performance optimization is described that improved the metric of (performance / (weight * volume * power)) by over a factor of 300 through the four generations.

The challenges of wearable computing: Part 1

“…Carnegie Mellon researchers have already demonstrated an English-to-Croatian wearable speech translator for use in the Bosnian peacekeeping mission. 4 Wearable computers can also help manage interruptions in the user's daily life. With the spread of cellular phones, more people are aware of the inconvenience of an untimely interruption.…”

Designs for a general purpose wearable computer

“…Wearable computing has been successfully applied in a wide variety of application domains including inspection, maintenance, manufacturing, navigation, remote collaboration, situational awareness, and language translation [15,16,17,18,19,20]. The areas of use are diverse but generally fall into one of three categories: application-specific use, general use, and context-aware use.…”

“…Much of the previous and current research uses tightly defined tasks to measure the results against [ 16,17,18]. In general, this is a good practice not only because it helps keeps the research focused but also because it provides an easy means for judging success or failure.…”