Digital integrated system for capturing, processing, and archiving of audio visual and data media content comprises an advanced and innovative system for capturing, processing, and archiving audio visual and data media content with support of existing analog and coming digital radio-diffusive technologies. The device delivers easy integration into supervision systems, processing and analytics of audio visual media content in real time or on the basis of post processing with local or dislocated archiving of captured audio visual or data media content including meta data
F.33 Slovenian patent
COBISS.SI-ID: 14456598Member of editorial board International journal of speech technology that publishes articles in the field of speech technology development and applications.
C.06 Editorial board membership
COBISS.SI-ID: 16846341The results of language analysis were used for developing Russian grapheme to phoneme conversion for Iskratel ASR automatic speech recognition system. The new module was introduced in the Iskratel ASR system. The support for voice driven telecommunication services with dynamic content was improved with the new module.
F.17 Transfer of existing technologies, know-how, methods and procedures into practice
COBISS.SI-ID: 14157846The analysis of accuracy of automatic speech driven service Autoattendant was carried out, based on the experimental results obtained in field trial test. The analysis encompased diffetent evaluation parameters important for evaluation of speech recognition accuracy, efficiency of error-recovery schemes, dialogue efficiency, user friendlines and service acceptability.
F.17 Transfer of existing technologies, know-how, methods and procedures into practice
COBISS.SI-ID: 14157590The work presents the steps for acquisition of speech material and steps for annotation of Slovenian Lombard Speech Database. The goal of this work was to record speech database with presence of Lombard effect in recorded speech material. Approximately 30 minutes of speech material per speaker and per session was recorded. The manual annotation of speech material was performed with the LombardSpeechLabel tool. The speech and annotation material was saved on 10 DVDs (one speaker on one DVD).
B.03 Paper at an international scientific conference
COBISS.SI-ID: 14145046