Speech and Language Resources within Speech Recognition and Synthesis Systems for Serbian and Kindred South Slavic Languages

Delić, Vlado; Seăujski, Milan; Jakovljević, Nikša; Pekar, Darko; Miskovic, Dragisa; Popović, Branislav M.; Ostrogonac, Stevan; Bojanić, Milana; Knežević, Milan

doi:10.1007/978-3-319-01931-4_42

Cited by 9 publications

(9 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Speech technologies for Serbian have been developed over the past decade and a half at the Faculty of Technical Sciences of the University of Novi Sad, in cooperation with the company “AlfaNum” from Novi Sad. During this period, a respectable amount of data for training ASR and TTS systems has been acquired [ 25 ], and both technologies are still being constantly improved by introducing new techniques and gathering new resources. The quality of ASR and TTS that is sufficient for most practical applications has been reached several years ago and, even though further research and development are needed in order to follow the state of the art, the research is now mainly focused on creating natural-like dialogue systems.…”

Section: Speech Technologies For Serbianmentioning

confidence: 99%

See 1 more Smart Citation

Educational Applications for Blind and Partially Sighted Pupils Based on Speech Technologies for Serbian

Lučić

Ostrogonac

Sedlar

et al. 2015

The Scientific World Journal

Self Cite

View full text Add to dashboard Cite

The inclusion of persons with disabilities has always represented an important issue. Advancements within the field of computer science have enabled the development of different types of aids, which have significantly improved the quality of life of the disabled. However, for some disabilities, such as visual impairment, the purpose of these aids is to establish an alternative communication channel and thus overcome the user's disability. Speech technologies play the crucial role in this process. This paper presents the ongoing efforts to create a set of educational applications based on speech technologies for Serbian for the early stages of education of blind and partially sighted children. Two educational applications dealing with memory exercises and comprehension of geometrical shapes are presented, along with the initial tests results obtained from research including visually impaired pupils.

show abstract

Section: Speech Technologies For Serbianmentioning

confidence: 99%

“…The feature vector, which defines the acoustic models of triphones, consists of 15 mel-frequency cepstral coefficients (MFCCs), normalized energy, and their derivatives. Feature vectors are extracted every 10 ms from 30 ms speech segments centered around extraction time instants [ 25 ].…”

Section: Speech Technologies For Serbianmentioning

confidence: 99%

Educational Applications for Blind and Partially Sighted Pupils Based on Speech Technologies for Serbian

Lučić

Ostrogonac

Sedlar

et al. 2015

The Scientific World Journal

Self Cite

View full text Add to dashboard Cite

show abstract

“…The feature vectors are extracted from 30 ms speech segments, every 10 ms. The training set for the acoustic model contains recordings of both scripted and spontaneous utterances produced by several dozen speakers, with a total duration of about 200 hours [23].…”

Section: Speech Recognitionmentioning

confidence: 99%

“…Our language model is a combination of 3 N-gram models. The first model is based on tokens (surface forms), the second on lemmata, and the third on classes [23]. The size of vocabulary causes data sparsity problems, resulting in the need for significantly greater language corpora, sufficient for obtaining a robust language model.…”

Section: Speech Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

User-awareness and adaptation in conversational agents

et al. 2014

View full text Add to dashboard Cite

This paper considers the research question of developing user-aware and adaptive conversational agents. The conversational agent is a system which is useraware to the extent that it recognizes the user identity and his/her emotional states that are relevant in a given interaction domain. The conversational agent is user-adaptive to the extent that it dynamically adapts its dialogue behavior according to the user and his/her emotional state. The paper summarizes some aspects of our previous work and presents work-in-progress in the field of speech-based human-machine interaction. It focuses particularly on the development of speech recognition modules in cooperation with both modules for emotion recognition and speaker recognition, as well as the dialogue management module. Finally, it proposes an architecture of a conversational agent that integrates those modules and improves each of them based on some kind of synergies among themselves.

show abstract