Stefano Fasciani

• Copiaco, Abigail; Ritz, Christian H; Fasciani, Stefano & Abdulaziz, Nidhal (2019). Scalogram Neural Network Activations with Machine Learning for Domestic Multi-channel Audio Classification. IEEE International Symposium on Signal Processing and Information Technology (ISSPIT).  ISSN 2162-7843.
• Fasciani, Stefano & Vohra, Manohar (2019). Towards a High-Performance Platform for Sonic Interaction Interfaces, In Isabel Barbancho; Lorenzo J. Tardón; Alberto Peinado & Ana M. Barbancho (ed.),  SMC 2019 Proceedings of the 16th Sound & Music Computing Conference.  Society for Sound and Music Computing.  ISBN 978-84-09-08518-7.  Paper.  s 97 - 104 Full text in Research Archive. Show summary

  In this paper we introduce a hardware platform to proto- type interfaces of demanding sonic interactive systems. We target applications featuring a large array of analog sensors requiring data acquisition and transmission to computers at fast rates, with low latency, and high band- width. This work is part of an ongoing project which aims to provide designers with a cost effective and acces- sible platform for fast prototyping of complex interfaces for sonic interactive systems or musical instruments. The high performances are guaranteed by a SoC FPGA. The functionality of the platform can be customized without requiring significant technical expertise. In this paper, we discuss the principles, the current design, and the prelim- inary evaluation against common microcontroller-based platforms. The proposed platform can sample up to 96 analog channels at rates up to 24 kHz and stream the data via UDP to computers with a sub millisecond latency.

• Vohra, Manohar & Fasciani, Stefano (2019). PYNQ-Torch: a framework to develop PyTorch accelerators on the PYNQ platform. IEEE International Symposium on Signal Processing and Information Technology (ISSPIT).  ISSN 2162-7843.
• Fasciani, Stefano (2018). Spectral Granular Synthesis, In Tae Hong Park & Doo-Jin Ahn (ed.),  Proceedings of the 2018 International Computer Music Conference.  The International Computer Music Association.  ISBN 0-9845274-7-8.  Paper.  s 99 - 103 Show summary

  This article introduces a granular synthesis algorithm in which grains of sound are processed in the frequency domain and combined at spectral level. In particular, each grain contributes to the sound synthesis with its magnitude spectrum only. Phase is reconstructed using spectrogram inversion techniques which support real- time computation. With this method, windowing and overlapping grains of regular size is no longer required. The algorithm presents a large number of parameters available for the dynamic manipulation of the synthetic timbre. Algorithm implementations are detailed in the paper, including comparison with traditional time-domain against the proposed frequency-domain granulation strategies.

• Fasciani, Stefano & Rahman, Habibur (2018). Tangible Virtual Patch Cords, In Tae Hong Park & Doo-Jin Ahn (ed.),  Proceedings of the 2018 International Computer Music Conference.  The International Computer Music Association.  ISBN 0-9845274-7-8.  Paper.  s 316 - 321 Show summary

  This paper presents a system to tangibly manipulate the virtual patching cords in graphical programming environments, such as Max and Pure Data. The system includes a physical interface, a communication protocol, and a software library, providing physical extension of the graphical programming paradigm. The interface includes a patch bay with connectors representing signal inlet and outlets from the programming environment. When inlets and outlets are connected with patching cords, equivalent virtual connections are created at runtime. The system supports one-to-many and many-to-one connections with different signal combination schemes. The design of the hardware and software components of the current prototype is detailed in the paper, as well as possible use of the system for programming and live performances.

• Fasciani, Stefano & Wyse, Lonce (2018). Vocal control of sound synthesis personalized by unsupervised machine listening and learning. Computer Music Journal.  ISSN 0148-9267.  42(1), s 37- 59 . doi: 10.1162/comj_a_00450 Show summary

  In this article we describe a user-driven adaptive method to control the sonic response of digital musical instruments using information extracted from the timbre of the human voice. The mapping between heterogeneous attributes of the input and output timbres is determined from data collected through machine-listening techniques and then processed by unsupervised machine-learning algorithms. This approach is based on a minimum-loss mapping that hides any synthesizer-specific parameters and that maps the vocal interaction directly to perceptual characteristics of the generated sound. The mapping adapts to the dynamics detected in the voice and maximizes the timbral space covered by the sound synthesizer. The strategies for mapping vocal control to perceptual timbral features and for automating the customization of vocal interfaces for different users and synthesizers, in general, are evaluated through a variety of qualitative and quantitative methods.

• Lutfi, Aoun & Fasciani, Stefano (2018). Automating the Optimization of Web Interfaces for E-Commerce, In Jihad Mohamad Alja’am; Abdulmotaleb El Saddik & Abdul Hamid Sadka (ed.),  Recent Trends in Computer Applications.  Springer Nature.  ISBN 978-3-319-89914-5.  Chapter.  s 221 - 235 Show summary

  In this chapter, we discuss a computational method that supports the design, revision, and amendment of web e-commerce GUI by finding the optimal placement of interactive elements. The definition of optimal placement is context-specific and it aims at maximizing measurable aspects of the user experience. We compute a visual model of an interface and we compare it against a set of design rules derived from a reference model. The generation of the GUI layout model is unsupervised and based only on image processing techniques. The model is generated exclusively using the browser’s rendering, which is visually presented to users. The comparison rates the GUI against the rules and provides a set of recommendations on how to modify the interface to increase the similarity with the reference model. When multiple reference models exist, we use a classifier to select the appropriate one. To prove this concept, we present a study on e-commerce websites where placement of the checkout button has a significant impact on the online sale process conversion rate. The system identifies nonoptimal placement, and it recommends an alternative position that is likely to improve the conversion rate. The system is implemented as an open-source software, and it currently supports the repositioning of a single interactive element. In the chapter, we discuss a further generalization of this approach to support a larger number of interactive objects in a wider spectrum of scenarios.

• Fasciani, Stefano (2017). Physical Audio Digital Filters. Proceedings of the International Conference on New Interfaces for Musical Expression.  ISSN 2220-4792.  s 476- 480 . doi: 10.5281/zenodo.1176322 Show summary

  We propose an approach to insert physical objects in audio digital signal processing chains, filtering the sound with the acoustic impulse response of any solid measured in real-time. We model physical objects as a linear time-invariant system, which is used as an audio filter. By interacting with the object or with the measuring hardware we can dynamically modify the characteristics of the filter. The impulse response is obtained correlating a noise signal injected in the object through an acoustic actuator with the signal received from an acoustic sensor placed on the object. We also present an efficient multichannel implementation of the system, which enables further creative applications beyond audio filtering, including tangible signal patching and sound spatialization.

• Lutfi, Aoun & Fasciani, Stefano (2017). Towards Automated Optimization of Web Interfaces and Application to E-commerce, In Mahmoud Al Qutayri; Jihad Mohamad Alja’am & Mohamed Eid (ed.),  2017 International Conference on Computer and Applications (ICCA).  IEEE.  ISBN 978-1-5386-2752-5.  Paper.  s 79 - 84 Show summary

  In this paper, we present a system that supports the design of web graphical user interface by finding the optimal placement of interactive elements. The definition of optimal placement is context specific; it aims at maximizing measurable aspects of the user experience, and it is derived using expert knowledge embedded in the system, which is based on HCI principles, user studies, and data analytics. We use a novel modeling technique to represent the layout of web user interfaces. The model is unsupervisedly computed on information extracted by several image processing algorithms. The system identifies the website category, builds a layout model, compares it with the relevant optimal model, and recommends an alternative layout of interactive elements. To prove this concept, we present a study on e-commerce websites where placement of the checkout button has a significant impact on the online sale process conversion rate. The system identifies non-optimal placement, and it recommends an alternative position that is likely to improve the conversion rate. The results will be validated based on the Visa Checkout user experience. The system is implemented as an open-source software, and it currently supports the re-positioning of a single interactive element. In the paper, we discuss a further generalization of this approach to support a larger number of interactive objects in a wider spectrum of scenarios.

View all works in Cristin

• Fasciani, Stefano; Jensenius, Alexander Refsum; Støckert, Robin & Xambó, Anna (2020). The MCT Portal: an infrastructure, a laboratory and a pedagogical tool.

View all works in Cristin