Loading...
Projects / Programmes source: ARIS

Advanced methods of interaction in telecommunication

Periods
Research activity

Code Science Field Subfield
2.08.00  Engineering sciences and technologies  Telecommunications   
1.07.01  Natural sciences and mathematics  Computer intensive methods and applications  Algorithms 

Code Science Field
T180  Technological sciences  Telecommunication engineering 

Code Science Field
2.02  Engineering and Technology  Electrical engineering, Electronic engineering, Information engineering 
1.01  Natural Sciences  Mathematics 
Evaluation (rules)
source: COBISS
Researchers (18)
no. Code Name and surname Research area Role Period No. of publicationsNo. of publications
1.  02071  PhD Zmago Brezočnik  Computer science and informatics  Researcher  2014 - 2017  456 
2.  33286  PhD Gregor Donaj  Telecommunications  Researcher  2014 - 2017  69 
3.  06821  PhD Zdravko Kačič  Telecommunications  Head  2014 - 2017  705 
4.  00873  PhD Tatjana Kapus  Computer science and informatics  Researcher  2014 - 2017  282 
5.  27566  PhD Marko Kos  Telecommunications  Researcher  2014 - 2017  103 
6.  18590  PhD Iztok Kramberger  Electronic components and technologies  Researcher  2014 - 2017  777 
7.  50324  PhD Izidor Mlakar  Telecommunications  Beginner researcher  2017  131 
8.  50218  PhD Grega Močnik  Telecommunications  Technical associate  2017  38 
9.  34283  Marko Presker    Technical associate  2014 - 2015 
10.  18876  PhD Matej Rojc  Telecommunications  Researcher  2014 - 2017  247 
11.  37698  PhD David Selčan  Electronic components and technologies  Junior researcher  2015 - 2017  34 
12.  18168  PhD Mirjam Sepesy Maučec  Telecommunications  Researcher  2014 - 2017  250 
13.  23838  PhD Darinka Verdonik  Linguistics  Researcher  2014 - 2017  200 
14.  22290  PhD Damjan Vlaj  Telecommunications  Researcher  2014 - 2017  165 
15.  20030  PhD Boštjan Vlaovič  Telecommunications  Researcher  2014 - 2017  195 
16.  22441  PhD Aleksander Vreže  Telecommunications  Researcher  2014  66 
17.  34282  Danilo Zimšek  Telecommunications  Technical associate  2014 - 2015  40 
18.  20032  PhD Andrej Žgank  Telecommunications  Researcher  2014 - 2017  241 
Organisations (1)
no. Code Research organisation City Registration number No. of publicationsNo. of publications
1.  0796  University of Maribor, Faculty of Electrical Engineering and Computer Science  Maribor  5089638003  27,550 
Abstract
The research work will be carried out within two research fields: multimodal interfaces and ensuring the system correctness. In the field of multimodal interfaces research will address the key issues, which will allow successful implementation of speech technologies for Slovenian language in modern communication environments, such as the concept of supportive living environments and intelligent ambience, as well as automated telecommunications services. The research will base on the results achieved on spontaneous speech recognition and will extend to conversational speech recognition. It will encompass developent of new algorithms for robust processing of speech in noise-intensive environments in the case of spatial voice recording and new ways of modeling and detecting spontaneous and coversational speech. Research will further include speech modeling using the processing of more complex structures, which are as important for achieving good performance speech recognition, as well as to engage in a variety of telecommunications services. Language modeling research will focus on the use of morpho-syntactic tags, which are important knowledge source for highly inflected languages. The importance of morpho-syntactic tags will be also investigated in the scope of the search for the language structure and content bearing word. In the field of speech-centered machine translation research will include the development of methods for statistical translation, taking into account the peculiarities of inflected languages and subtitling of video content. The research will also include development of dialog management models for spoken dialogue systems. The work in the framework of system correctness assurance will include theoretical, development, and applicative research in the area of system design with the help of formal methods. We will investigate the methods of specification by using different kinds of automata, temporal logics, and standardised specification languages in order to be able to effectively assure the logical correctness and check real-time and probabilistic properties of systems. The subject of this research programme are the systems composed of many components which execute concurrently and continuously communicate with one another and with the environment. The examples thereof are communications protocols and multithreaded programs. We will investigate the relation between the classical methods for system analysis and formal verification methods. The verification will mainly be carried out by the model checking method with the help of our own and other tools, as this method checks the system state space automatically and does not require much interaction with the user. In order for the formal verification to be more user-friendly for its use in industry, we will supplement our development environment SpinRCP for the well-known Spin verification tool by additional features.
Significance for science
Convergence of information and communication technologies defines new communication contexts, which poses new challenges and sets new demanding tasks to the field of language technologies, and at the same time opens new opportunities for speech technologies to bring significant added value to new telecommunication services. With development of the concepts of supportive living environments, intelligent ambience, and automated teleservices, the speech technologies take the role of essential enabling technologies that will significantly contribute to development of next generation communication environments foreseen in various development strategies, among others also in the strategic documents of the Horizon 2020. Here, the language dependent nature of language technologies is an important issue, as the technology is not easily transferable to other languages. Therefore, the development of language-specific technological solutions for such systems is even more important. The topic of spontaneous speech recognition used in telecommunication services is being intensively researched worldwide. Each language brings some new peculiarities, not previously covered with general technological solutions. We expect that our research will bring an important contribution to the development of technologies for highly inflectional languages. The field of spontaneous and conversational speech processing, especially in the light of automatic speech recognition, still poses many open questions. The research in this field will foster development of advanced and more complex technologies for Slovenian language, especially in the context of spoken dialogue systems. Building language resources and tools, be it either known resources and tools with additional quality or  new types of resources and tools, represents a crucial link through which the knowledge from linguistics is transferred to technologies. The research in the area of real-time and probabilistic verification will contribute to a better knowledge of the capabilities of existing methods and tools and potentially to the improvement of the tools developed by the groups abroad. The use of the formal methods instead of the classical ones will also allow for checking the validity of the existing results of the analysis of selected communications systems carried out by using the classical approach. The use of formal methods also enables a better insight into the system functioning, which potentially can lead to ideas for improvement of the systems or a better selection of the working parameters. After our Spin plug-in in Eclipse, the enhanced and with new capabilities complemented SpinRCP development environment is our next product that will be offered to the developers and users of the well-known Spin verification tool. Thereby we will contribute to the development of science and greater use of formal methods in practice.
Significance for the country
Research work in speech and language technologies is of crucial importance for Slovenian language. Language dependency of speech technology makes the research in this field specific in a sense that it requires development of language specific solutions. The global technology providers provide various speech technology solutions for commercially more interesting languages, like English, Chinese, Spanish, French, German, Arabic, but few or non for languages that are spoken by smaller populations. Slovenian and many other highly inflected languages belong to the latter group of languages. Development of speech and language technologies for Slovenian language makes the language attractive for its integration into the state-of-the-art communication technology systems. Research results obtained for Slovenian language could be adapted to other inflective languages and to some extend to agglutinative languages. From this aspect the importance of research work expands over national borders. This fact is also important from the economic point of view, as potential market of many Slovenian companies includes countries, where languages of these two types are spoken. New language resources can have a socio-economic impact not merely through technologies which are developed based on them, but also directly, as is the case of the Slovenian reference speech corpus GOS, which is also an important tool for everyday user, and at the same time represents a valuable cultural inheritance for future generations. The results of proposed research are of direct relevance for Slovenian telecommunications industry, especially in the segment of automated telecommunication services development. The results are also important for audiovisual content providers, especially for the purposes of indexing and searching the audiovisual content (TV and radio stations, corporate media monitoring and analysis), for the operators - providing automated voice services, etc. With the development of assistive living and intelligent environments, as well as telecommunication services, the potential of voice technologies will substantially increase, which will widen the range of interested companies. An important contribution is also expected in the socio-economic field. The proposed research will contribute to increased inclusion of people with disabilities. Slovenian producers of ICT equipment must compete in the global market with large foreign companies that invest considerable resources in ensuring the high quality of their products and include the use of formal methods in the key stages of the development, as such an approach reduces the likelihood that subtle errors are discovered in a product no sooner than when being sold. By research in the field of formal methods we will enable also our companies to become aware of the potential of formal methods, and to incorporate them into their development processes, thereby reducing the cost of late discovered errors in their products.
Most important scientific results Annual report 2014, 2015, 2016, final report
Most important socioeconomically and culturally relevant results Annual report 2014, 2015, 2016, final report
Views history
Favourite