Model adaptation techniques are an efficient way to reduce the mismatch that typically occurs between the training and test condition of any speech recognizer. Adaptation techniques can usually be divided into two families of approaches. On one hand, direct model adaptation attempts to directly reestimate the model parameters, for example using MAP adaptation. Since direct adaptation only reestimates model parameters of the corresponding units appearing in the adaptation data, a large amount of such data is needed to observe any significant improvement in performance. However, nice asymptotic properties are usually observed, meaning that the performance improves as the amount of adaptation data increases. On the other hand, indirect model adaptation applies a general transformation on some clusters of model parameters. Because each individual model is transformed, the approach is quite effective when a small amount of adaptation data is available. However, as the amount of adaptation data increases, the performance improvement quickly saturates. In this paper, we propose to jointly estimate model parameters and transformation parameters using a single estimation criterion based on Bayesian statistics. We show that by providing a prior distribution for the model parameters and the transformation parameters, it is possible to jointly estimate these two sets of parameters using maximum a posteriori estimation (MAP). Experimental evaluation on nonnative speaker and channel adaptation illustrates the effectiveness of the proposed approach.
Software is increasingly individualized to the needs of customers and may have to be adapted to changing contexts and environments after deployment. Therefore, individualized software adaptations may have to be performed. As a large number of variants for aected systems and domains may exist, the creation and deployment of the individualized software should be performed automatically based on the software's conguration and context. In this paper, we present a toolchain to develop and deploy individualized software adaptations based on Software Product Line (SPL) engineering. In particular, we contribute a description and technical realization of a toolchain ranging from variability modeling over variability realization to variant derivation for the automated deployment of individualized software adaptations. To capture the variability within realization artifacts, we employ delta modeling, a transformational SPL implementation approach. As we aim to fulll requirements of industrial practice, we employ model-driven engineering using statecharts as realization artifacts. Particular statechart variants are further processed by generating C/C++ code, linking to external code artifacts, compiling and deploying to the target device. To allow for exible and parallel execution the toolchain is provided within a cloud environment. This way, required variants can automatically be created and deployed to target devices. We show the feasibility of our toolchain by developing the industry-related case of emergency response systems.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.