Pregled bibliografske jedinice broj: 595659
Controlling dynamic stochastic synthesis with an audio signal
Controlling dynamic stochastic synthesis with an audio signal // Proceedings of the International Computer Music Conference 2012 / Marolt, Kaltenbrunner, Ciglar (ur.).
San Francisco (CA): The International Computer Music Association, 2012. str. 100-104 (poster, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 595659 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Controlling dynamic stochastic synthesis with an audio signal
Autori
Kreković, Gordan ; Brkić, Igor
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the International Computer Music Conference 2012
/ Marolt, Kaltenbrunner, Ciglar - San Francisco (CA) : The International Computer Music Association, 2012, 100-104
ISBN
978-0-9845274-1-0
Skup
International Computer Music Conference 2012
Mjesto i datum
Ljubljana, Slovenija, 09.09.2012. - 14.09.2012
Vrsta sudjelovanja
Poster
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
dynamic stochastic synthesis; audio features; sound synthesis
Sažetak
Dynamic stochastic synthesis is a sound synthesis technique devised by Ianis Xenakis. This technique produces waveforms by interpolating a set of breakpoints. At every repetition of the waveform, the breakpoints change their positions accordingly to random walks. In all the previous implementations of this method, synthesis parameters were either fixed or controllable through a graphical user interface. Our intention was to hide most of the parameters from musicians and let them use an input audio signal to intuitively control the synthesis process. We proposed and implemented a novel solution based on a mapping between audio features of an input signal and the synthesis parameters. The amplitude random walk is limited by the minimum and maximum amplitudes of the input signal and parameterized by the scaled spectral entropy. To define the frequency range of stochastic waveforms, the algorithm uses the fundamental frequency for periodic input signals and the spectral centroid for noisy signals. It is possible to change the interpolation method and the number of breakpoints in a waveform using the graphical user interface. The solution was developed as a real-time system and demonstrated to professional composers. They confirmed that the realization is suitable for practical tasks, especially for live performances.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb
Profili:
GORDAN KREKOVIĆ
(autor)