2018 IEEE Spoken Language Technology Workshop (SLT)
Recently, deep learning methods have been increasingly applied on spoken language technologies, including signal processing, language understanding and generation, dialogue management, as well as joint optimisations of these (end-to-end learning). However, such methods still have limitations and it is not yet clear that deep learning and joint optimisation is the key to the future.
Encompassing the current deep learning trends and traditional knowledge-based methods, SLT’s 2018 main theme will be around “Spoken Language Technology in the Era of Deep Learning: Challenges and Opportunities”.
Automatic assessment of spoken language proficiency is a sought-after technology. These systems often need to handle the operating scenario where candidates have a skill level or first language which was not encountered during the training stage. For high stakes tests it is necessary for those systems to have good grading performance when the candidate is from the same population as those contained in the training set, and they should know when they are likely to perform badly in the case when the candidate is not from the same population as the ones contained in training set. This paper focuses on using Deep Density Networks to yield auto-marking confidence. Firstly, we explore the benefits of parametrising either a predictive distribution or a posterior distribution over the parameters of the model likelihood and obtaining the predictive distribution via marginalisation. Secondly, we investigate how it is possible to act on the parametrised density in order to explicitly teach the model to have low confidence in areas of the observation space where there is no training data by assigning confidence scores to artificially generated data. Lastly, we compare the capabilities of Factor Analysis, Variational Auto-Encodes, and Wasserstein Generative Adversarial Networks to generate artificial data.
This paper addresses the problem of stylized text generation in a multilingual setup. A version of a language model based on a long short-term memory (LSTM) artificial neural network with extended phonetic and semantic embeddings is used for stylized poetry generation. The quality of the resulting poems generated by the network is estimated through bilingual evaluation understudy (BLEU), a survey and a new cross-entropy based metric that is suggested for the problems of such type. The experiments show that the proposed model consistently outperforms random sample and vanilla-LSTM baselines, humans also tend to associate machine generated texts with the target author.