Extracting the prosodic information for Turkish broadcast news data and using on the sentence segmentation task
| Autor: | Dogan Dalva, Umit Guz, Izel D. Revidi, Hakan Gurkan |
|---|---|
| Přispěvatelé: | Bölüm Yok, Guz, Umit -- 0000-0002-4597-0954, [Dalva, Dogan -- Revidi, Izel D. -- Guz, Umit -- Gurkan, Hakan] Isik Univ, Elekt Elekt Muhendisligi Bolumu, Istanbul, Turkey |
| Rok vydání: | 2014 |
| Předmět: |
prosodic feature set
automatic speech segmentation Computer science Turkish business.industry Speech recognition Feature extraction computer.software_genre language.human_language ComputingMethodologies_PATTERNRECOGNITION prosody language sentence segmentation Entropy (information theory) NIST Artificial intelligence Hidden Markov model Sentence segmentation Prosody business computer Natural language processing |
| Zdroj: | SIU |
| Popis: | 22nd IEEE Signal Processing and Communications Applications Conference (SIU) -- APR 23-25, 2014 -- Karadeniz Teknik Univ, Trabzon, TURKEY WOS: 000356351400432 In this study, extracting the prosodic information for Turkish Broadcast News Data using the open source tools and comparing the sentence segmentation performances of these grouped prosodic information on the raw data obtained as an output from the Automatic Speech Recognition System are established. Especially for the sentence segmentation task, a very promising prosodic feature set is obtained. IEEE, Karadeniz Tech Univ, Dept Comp Engn & Elect & Elect Engn |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|