In this paper, we propose a novel discriminative language model, which can be applied quite generally. Compared to the well known N-gram language models, discriminative language models can achieve more accurate discrimination because they can employ overlapping features and nonlocal information. However, discriminative language models have been used only for re-ranking in specific applications ...

Metaphors suffuse language and affect how people think. A meta-analysis of metaphor framing studies conducted between 1983 and 2000 concluded that metaphors are about 6% more persuasive than literal language (Sopory & Dillard, 2002). However, each of these studies was conducted in English with samples drawn from populations of native English speakers. Here, we test whether and how language prof...

We investigate the characteristics and quantifiable predispositions of both n-gram and recurrent neural language models in the framework of language generation. In modern applications, neural models have been widely adopted, as they have empirically provided better results. However, there is a lack of deep analysis of the models and how they relate to real language and its structural properties...

In order to understand how language abilities emerge in typically and atypically developing infants and toddlers, it is important to embrace complexity in development. In this paper, we describe evidence that early language development is an experience-dependent process, shaped by diverse, interconnected, interdependent developmental mechanisms, processes, and abilities (e.g. statistical learni...

For several years, the Interspeech ComParE Challenge has focused on paralinguistic tasks of various kinds. In this paper we focus on the Native Language and the Deception subchallenges of ComParE 2016, where the goal is to identify the native language of the speaker, and to recognize deceptive speech. As both tasks can be treated as classification ones, we experiment with several state-of-the-a...

The present article provides a comprehensive review of the work carried out on developing PRESEMT, a hybrid language-independent machine translation (MT) methodology. This methodology has been designed to facilitate rapid creation of MT systems for unconstrained language pairs, setting the lowest possible requirements on specialised resources and tools. Given the limited availability of resourc...

We investigate data driven natural language generation under the constraints that all words must come from a fixed vocabulary and a specified word must appear in the generated sentence, motivated by the possibility for automatic generation of language education exercises. We present fast and accurate approximations to the ideal rejection samplers for these constraints and compare various senten...

This paper introduces a new confidence scoring mechanism which takes advantage of the semantic parsing provided by a natural language understanding system. As a result, different segments of the user input which fill individual semantic slots are identified and individual semantic slot confidence scores are generated. With slot-based confidence scores, each individual slot receives a confidence...