Domain-specific languages (DSLs) provide high-level and domain-specific abstractions that allow expressive and concise algorithm descriptions. Since the description in a DSL hides also the properties of the target hardware, DSLs are a promising path to target different parallel and heterogeneous hardware from the same algorithm description. In theory, the DSL description can capture all characteristics of the algorithm that are required to generate highly efficient parallel implementations. However, most frameworks do not make use of this knowledge and the performance cannot reach that of optimized library implementations. In this article, we present the HIPAccframework, a DSL and source-to-source compiler for image processing. We show that domain knowledge can be captured in the language and that this knowledge enables us to generate tailored implementations for a given target architecture. Back ends for CUDA, OpenCL, and Renderscript allow us to target discrete graphics processing units (GPUs) as well as mobile, embedded GPUs. Exploiting the captured domain knowledge, we can generate specialized algorithm variants that reach the maximal achievable performance due to the peak memory bandwidth. These implementations outperform state-of-the-art domain-specific languages and libraries significantly.

To cope with the complexity of programming GPU accelerators for medical imaging computations, we developed a framework to describe image processing kernels in a domain-specific language, which is embedded into C++. The description uses decoupled access/execute metadata, which allow the programmer to specify both execution constraints and memory access patterns of kernels. A source-to-source compiler translates this high-level description into low-level CUDA and Open CL code with automatic support for boundary handling and filter masks. Taking the annotated metadata and the characteristics of the parallel GPU execution model into account, two-layered parallel implementations - utilizing SPMD and MPMD parallelism - are generated. An abstract hardware model of graphics card architectures allows to model GPUs of multiple vendors like AMD and NVIDIA, and to generate device-specific code for multiple targets. It is shown that the generated code is faster than manual implementations and those relying on hardware support for boundary handling. Implementations from Rapid Mind, a commercial framework for GPU programming, are outperformed and similar results achieved compared to the GPU backend of the widely used image processing library Open CV.

An efficient memory bandwidth utilization for GPU accelerators is crucial for memory bound applications. In medical imaging, the performance of many kernels is limited by the available memory bandwidth since only a few operations are performed per pixel. For such kernels only a fraction of the compute power provided by GPU accelerators can be exploited and performance is predetermined by memory bandwidth. As a remedy, this paper investigates the optimal utilization of available memory bandwidth by means of increasing in-flight memory transactions. Instead of doing this manually for different GPU accelerators, the required CUDA and OpenCL code is automatically generated from descriptions in a Domain-Specific Language (DSL) for the considered application domain. Moreover, the DSL is extended to also support global reduction operators. We show that the generated target-specific code improves bandwidth utilization for memory-bound kernels significantly. Moreover, competitive performance compared to the GPU back end of the widely used image processing library OpenCV can be achieved.

The design and implementation of the AT&T Communicator mixed-initiative spoken dialog system is described.The Communicator project, sponsored by DARPA and launched in 1999, is a multi-year multi-site project on advanced spoken dialog systems research.The main focus of this paper is on issues related to the design of mixed-initiative systems.In addition to describing our architecture and implementation of the complex travel task, the paper reports on some preliminary evaluation results.

This paper describes our progress towards building a telephony-based spoken dialogue system for workshop/conference services. A mixed-initiative dialogue system has been developed that is engineered to o er users natural interaction with the system, ease-of-use and robustness towards ambiguous requests and machine errors. A prototype system, known as W99, is described in this paper

We propose a quantitative model for dialog systems that can be used for learning the dialog strategy. We claim that the problem of dialog design can be formalized as an optimization problem with an objective function reflecting different dialog dimensions relevant for a given application. We also show that any dialog system can be formally described as a sequential decision process in terms of its state space, action set, and strategy. With additional assumptions about the state transition probabilities and cost assignment, a dialog system can be mapped to a stochastic model known as Markov decision process (MDP). A variety of data driven algorithms for finding the optimal strategy (i.e., the one that optimizes the criterion) is available within the MDP framework, based on reinforcement learning. For an effective use of the available training data we propose a combination of supervised and reinforcement learning: the supervised learning is used to estimate a model of the user, i.e., the MDP parameters that quantify the user's behavior. Then a reinforcement learning algorithm is used to estimate the optimal strategy while the system interacts with the simulated user. This approach is tested for learning the strategy in an air travel information system (ATIS) task. The experimental results we present in this paper show that it is indeed possible to find a simple criterion, a state space representation, and a simulated user parameterization in order to automatically learn a relatively complex dialog behavior, similar to one that was heuristically designed by several research groups.

We introduce a stochastic model for dialogue systems based on Markov decision process. Within this framework we show that the problem of dialogue strategy design can be stated as an optimization problem, and solved by a variety of methods, including the reinforcement learning approach. The advantages of this new paradigm include objective evaluation of dialogue systems and their automatic design and adaptation. We show some preliminary results on learning a dialogue strategy for an air travel information system.

We present a new approach of combining stochastic language models and traditional linguistic models to enhance the performance of our spontaneous speech recognizer. We compile arbitrary large linguistic context dependencies into a category based bigram model which allows us to use a standard beam-search driven forward Viterbi algorithm for real time decoding. Since this recognizer is used in a dialog system, the information about the last system utterance is used to build dialogstep dependent language models. This setup is verified and tested on our corpus of spontaneous speech utterances collected with our dialog system. Experimental results show a significant reduction of word error rate.

In this paper we describe an approach to automatic evaluation of both the speech recognition and understanding capabilities of a spoken dialogue system for train time table information. We use word accuracy for recognition and concept accuracy for understanding performance judgement. Both measures are calculated by comparing these modules' output with a correct reference answer. We report evaluation results for a spontaneous speech corpus with about 10000 utterances. We observed a nearly linear relationship between word accuracy and concept accuracy.

In this paper we describe the semantic interpretation process of utterances in a spoken dialog system for train table inquiries. Spoken dialogs show a large set of prob-lems in human–machine–communication like stops, cor-rections, filled pauses, non grammatical sentences, el-lipses, unconnected phrases etc. In our robust approach we are able to handle a substantial amount of them. While the principles of robustness are shared in several mod-ules, in this paper we concentrate on the aspect of robust semantic analysis and domain specific interpretation of spoken utterances.

This paper focuses on the evaluation of the German Sundial Demonstrator maintaining interactive conversations via microphone and telephone with users. The word recognizer was implemented by the University of Erlangen and currently obtains on our test set a word accuracy of over 92% in a speaker--independent task with perplexity 111. We also participated in the design and implementation of the multilingual Dialogue Manager which is responsible for cooperative system behaviour. Over 100 prototype dialogues led to the current version which is continually being extended. The overall system performance is tested with semi-- naive users. Each subject faces four intercity train timetable scenarios, two of them are given, the others depend on the subjects' personal choice. Global system evaluation is done according to performance measures like contextual appropriateness, transaction success and dialogue completion rate. 1 Introduction Many systems providing human computer interaction have b...

This paper presents an approach to managing spoken dialogues in information services systems. We describe how the approach, based upon a tri--partite model of interaction, addresses the problems of co--operativeness and portability across languages and task domains. This approach has been implemented in the generic dialogue manager of the Sundial dialogue systems. We outline the capabilities of this dialogue manager and describe our testing methodology. Keywords: Oral Dialogue, Dialogue Management, Interpretation of Utterances, Cooperative System 1 Introduction Management of spoken dialogues requires the interpretation of the user utterances and the generation of system utterances coherent with user utterances. The underlying model of interaction is partitioned into semantic, task and dialogue models (cf. [4]). Utterances in different languages are mapped into a common semantic representation language from which instances of the semantic, task and dialogue models are constructed. Se...

The authors address the choice of suitable subword units for the hidden Markov model (HMM)-based front-end of a speaker-independent large vocabulary continuous speech dialog system (EVAR). In contrast to the well-known approach of using context-dependent phone-like units (for instance generalized triphones) the authors developed inventories of larger-sized subword units, so-called context-freezing units (CFU). CFU models can be considered as an approximation to the extremely desirable situation of having whole word HMMs under the limiting conditions of the training speech data at hand. Recognition experiments indicate an advantage of the context-freezing units over triphone/biphone/phone combinations in terms of the achieved word accuracy, at least in the case of German speech. Using triphones with contexts generalized by means of broad phonetic classes, the authors achieved results comparable to the CFU ones.>