No abstract

In this research, a study of cross-linguistic speech emotion recognition is performed. For this purpose, emotional data of different languages (English, Lithuanian, German, Spanish, Serbian, and Polish) are collected, resulting in a cross-linguistic speech emotion dataset with the size of more than 10.000 emotional utterances. Despite the bi-modal character of the databases gathered, our focus is on the acoustic representation only. The assumption is that the speech audio signal carries sufficient emotional information to detect and retrieve it. Several two-dimensional acoustic feature spaces, such as cochleagrams, spectrograms, mel-cepstrograms, and fractal dimension-based space, are employed as the representations of speech emotional features. A convolutional neural network (CNN) is used as a classifier. The results show the superiority of cochleagrams over other feature spaces utilized. In the CNN-based speaker-independent cross-linguistic speech emotion recognition (SER) experiment, the accuracy of over 90% is achieved, which is close to the monolingual case of SER.

Developing artificial learning systems that can understand and generate natural language has been one of the long-standing goals of artificial intelligence. Recent decades have witnessed an impressive progress on both of these problems, giving rise to a new family of approaches. Especially, the advances in deep learning over the past couple of years have led to neural approaches to natural language generation (NLG). These methods combine generative language learning techniques with neural-networks based frameworks. With a wide range of applications in natural language processing, neural NLG (NNLG) is a new and fast growing field of research. In this state-of-the-art report, we investigate the recent developments and applications of NNLG in its full extent from a multidimensional view, covering critical perspectives such as multimodality, multilinguality, controllability and learning strategies. We summarize the fundamental building blocks of NNLG approaches from these aspects and provide detailed reviews of commonly used preprocessing steps and basic neural architectures. This report also focuses on the seminal applications of these NNLG models such as machine translation, description generation, automatic speech recognition, abstractive summarization, text simplification, question answering and generation, and dialogue generation. Finally, we conclude with a thorough discussion of the described frameworks by pointing out some open research directions.

The recent introduction of whole-slide scanning systems enabled accumulation of highquality pathology images into large collections, thus opening new perspectives in cancer research, as well as new analysis challenges. Automated identification of tumour tissue in the whole-slide image enables further use of developed grading systems that classify tumour cell abnormalities and predict tumour developments. In this article, we describe several possibilities to achieve epithelium-stroma classification of tumour tissues in digital pathology images by employing annotated superpixels to train machine learning algorithms. We emphasize that annotating superpixels rather than manually outlining tissue classes in raw images is less time consuming, and more effective way of producing ground truth for computational pathology pipelines. In our approach feature space for supervised learning is created from tissue class assigned superpixels by extracting colour and texture parameters, and applying dimensionality reduction methods. Alternatively, to train convolutional neural network, labelled superpixels are used to generate square image patches by moving fixed size window around each superpixel centroid. The proposed method simplifies the process of ground truth data collection and should minimize the time spent by a skilled expert to perform manual annotation of whole-slide images. We evaluate our method on a private data set of colorectal cancer images. Obtained results confirm that a method produces accurate reference data suitable for the use of different machine learning based classification algorithms.

Automatic classification methods, such as artificial neural networks (ANNs), the k-nearest neighbor (kNN) and self-organizing maps (SOMs), are applied to allophone analysis based on recorded speech. A list of 650 words was created for that purpose, containing positionally and/or contextually conditioned allophones. For each word, a group of 16 native and non-native speakers were audio-video recorded, from which seven native speakers’ and phonology experts’ speech was selected for analyses. For the purpose of the present study, a sub-list of 103 words containing the English alveolar lateral phoneme /l/ was compiled. The list includes ‘dark’ (velarized) allophonic realizations (which occur before a consonant or at the end of the word before silence) and 52 ‘clear’ allophonic realizations (which occur before a vowel), as well as voicing variants. The recorded signals were segmented into allophones and parametrized using a set of descriptors, originating from the MPEG 7 standard, plus dedicated time-based parameters as well as modified MFCC features proposed by the authors. Classification methods such as ANNs, the kNN and the SOM were employed to automatically detect the two types of allophones. Various sets of features were tested to achieve the best performance of the automatic methods. In the final experiment, a selected set of features was used for automatic evaluation of the pronunciation of dark /l/ by non-native speakers.