Acoustic features determination for regularity articulation quantification of Serbian fricatives
| Autor: | Slobodan T. Jovicic, Silvana Punišić, Drasko Furundzic, Miško Subotić |
|---|---|
| Rok vydání: | 2012 |
| Předmět: |
Artificial neural network
Computer science business.industry Quality assessment Speech recognition SIGNAL (programming language) 020206 networking & telecommunications 02 engineering and technology Pronunciation computer.software_genre language.human_language 030507 speech-language pathology & audiology 03 medical and health sciences 0202 electrical engineering electronic engineering information engineering language Artificial intelligence 0305 other medical science business Serbian Articulation (phonetics) computer Natural language processing |
| Zdroj: | 11th Symposium on Neural Network Applications in Electrical Engineering. |
| DOI: | 10.1109/neurel.2012.6420008 |
| Popis: | this paper presents a brief methodology for determination of relevant parameters for automatic articulation regularity quantification of Serbian fricatives pronunciation. Using the known methods and algorithms as well as developing new ones for analysis of phoneme pronunciation process, authors used neural networks as a tool for modeling. Here we present a comparison of fricative articulation quality assessment, based on the different features of the speech signal. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|