The TeamTalk Corpus: Route Instructions in Open Spaces
نویسندگان
چکیده
This paper describes the TeamTalk corpus, a new corpus of route instructions consisting of directions given to a robot. Participants provided instructions to a robot that needed to move to a marked location. The environment contained two robots and a symbolic destination marker, all within an open space. The corpus contains the collected speech, speech transcriptions, stimuli, and logs of all participant interactions from the experiment. Route instruction transcriptions are divided into steps and annotated as either metric-based or landmarkbased instructions. This corpus captured variability in directions for robots represented in 2-dimensional schematic, 3-dimensional virtual, and natural environments, all in the context of open space navigation.
منابع مشابه
TeamTalk: A Platform for Multi-Human-Robot Dialog Research in Coherent Real and Virtual Spaces
Performing experiments with human-robot interfaces often requires the allocation of expensive and complex hardware and large physical spaces. Those costs constrain development and research to the currently affordable resources, and they retard the testing-andredevelopment cycle. In order to explore research free from mundane allocation constraints and speed-up our platform development cycle, we...
متن کاملWalk the Talk: Connecting Language, Knowledge, Action in Route Instructions
Following verbal route instructions requires knowledge of language, space, action and perception. We present MARCO, an agent that follows free-form, natural language route instructions by representing and executing a sequence of compound action specifications that model which actions to take under which conditions. MARCO infers implicit actions from knowledge of both linguistic conditional phra...
متن کاملWalk the Talk: Connecting Language, Knowledge, and Action in Route Instructions
Following verbal route instructions requires knowledge of language, space, action and perception. We present MARCO, an agent that follows free-form, natural language route instructions by representing and executing a sequence of compound action specifications that model which actions to take under which conditions. MARCO infers implicit actions from knowledge of both linguistic conditional phra...
متن کاملLearning to Follow Navigational Route Instructions
We have developed a simulation model that accepts instructions in unconstrained natural language, and then guides a robot to the correct destination. The instructions are segmented on the basis of the actions to be taken, and each segment is labeled with the required action. This flat formulation reduces the problem to a sequential labeling task, to which machine learning methods are applied. W...
متن کاملThe Structure and Generality of Spoken Route Instructions
A robust system that understands route instructions should be able to process instructions generated naturally by humans. Also desirable would be the ability to handle repairs and other modifications to existing instructions. To this end, we collected a corpus of spoken instructions (and modified instructions) produced by subjects provided with an origin and a destination. We found that instruc...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2015