We introduce a language generation dataset grounded in a popular video game.KNUDGE (KNowledge Constrained User-NPC Dialogue GEneration) requires models to produce trees of dialogue between video game characters that accurately reflect quest and entity specifications stated in natural language.KNUDGE is constructed from side quest dialogues drawn directly from game data of Obsidian Entertainment's The Outer Worlds, leading to real-world complexities in generation: (1) utterances must remain faithful to the game lore, including character personas and backstories; (2) a dialogue must accurately reveal new quest details to the human player; and (3) dialogues are large trees as opposed to linear chains of utterances.We report results for a set of neural generation models using supervised and in-context learning techniques; we find competent performance but room for future work addressing the challenges of creating realistic, game-quality dialogues.

An ultrawide-band radar system was developed to measure snow thickness from an airborne platform. It is a second-generation device, improving upon a system that was deployed to Antarctica in 2018-2019. The radar operates from 2.7 -10.7 GHz and from 10-18 GHz for a theoretical resolution of less than 2 cm, It was deployed to the Grand Mesa National Forest in Colorado on a Twin Otter DHC-6 aircraft in March of 2019. The system successfully mapped the air-snow and snow-ground interfaces as well as some internal layering over both vegetative and non-vegetative areas. In this paper, we discuss the design, development and deployment of this radar and show some preliminary results.

Accurate on-device wake word detection is crucial to products with far-field voice control such as the Amazon Echo. It is quite challenging to build a wake word system with both low False Reject Rate (FRR) and low False Alarm Rate (FAR) in real scenarios where there are various types of background speech, music or noise, especially when computational resources on the device is limited. In this paper, we introduce a two-stage wake word system based on Deep Neural Network (DNN) acoustic modeling, propose a new way to model the non-keyword background events using monophone-based units and present how richer information can be extracted from those monophone units for final wake word detection. Under the new system, we could get around 16% relative reduction in FRR when fixing the false alarm level, and about 37% relative reduction in FAR on the other hand if we maintain the miss rate. For the 2nd stage classifier itself, it is able to reduce the false alarm rate relatively by about 67% on top of 1st stage hypothesis with very few computational resources.