RessoaSom: Concrete reuse of sounds from the archive of Reverbera! project

RessoaSom: Concrete reuse of sounds from the archive of Reverbera! project Ariane de Souza Stolfi Universidade Federal do Sul da Bahia [email protected] Rafael Portugal Oliveira Universidade Federal do Sul da Bahia [email protected] Lucas Oliveira Rosário Universidade Federal do Sul da Bahia [email protected] ABSTRACT This project aims to organize and make available a repository of experimental sounds for the creation of an open sound library to be used in creative, educational, and professional productions. Drawing from an extensive archive of recordings made during rehearsals and performances involving free improvisation practices produced since 2024 by the group Reverbera!, the material was edited, analysed, segmented, and categorized based on morphological criteria. In addition to releasing the sounds under Creative Commons licenses, experimental digital interfaces were also developed to enable live performance with these sounds. RessonaSom is an experimental platform for live performance, where users can play some of the project’s mapped sounds in a random way, but guided by morphological criteria. 1. INTRODUCTION The Ressoa Project aims to produce a repository of free experimental sounds to be reused in creative, educational, or professional productions, both within and beyond the university. As a starting point, we drew upon a collection of recordings produced over the past year by the Reverbera! project — an outreach initiative that has been developed since 2018 as a space for experimentation in free musical improvisation practices. Our work is situated within the broader field of experimental music, a creative domain that breaks away from conventional standards of composition, performance, and listening. Experimental music is not defined by a specific style, but by a questioning attitude toward traditional musical codes, exploring new ways of producing sound [8]. With this project, we aim to expand the repertoire of experimental sounds available in public sound libraries, offering materials that can be used in a variety of artistic and cultural works. 2. REVERBERA! The project began in 2018 as a subproject of Imagina! Circuito Permanente de Audiovisual [4], and was interrupted by the COVID-19 pandemic. In 2023, it was resumed as an Licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0). Attribution: owner/author(s). Web Audio Conference WAC-2025, November 19–21, 2025, Paris, France. © 2025 Copyright held by the owner/author(s). Figure 1: Reverbera’s project team. Picture: Ariane Stolfi independent project with the participation of five students. In 2024, the project expanded to include eight students and a professional musician as performers, as well as a journalism student collaborating with coverage on our social media platforms. The current team has a diverse profile, including a majority of black people (figure 1). From the beginning, we have used classic silent films as a starting point for real-time sound experimentation through free improvisation practices. In recent years, we have presented films with live soundtracks for works by various filmmakers such as Georges M´eli`es, Ladislas Starevich, Salvador Dal´ı, Man Ray, Maya Deren, among others. Some of the films with live soundtracks that we presented last year can be viewed on our YouTube channel: https://www.youtube.com/@Reverbera.imagina. In his book Audio-Vision, Chion [3] highlights the importance of sound in audiovisual media, emphasizing that it can add a range of expressive values to the moving images on screen — particularly through music, which enhances the emotional tone of situations portrayed. In our performances, we aim to deepen the audience’s engagement with the films by providing a sonic dimension that encourages immersive experience. Since July 2024, we have also received support from the Paulo Gustavo Law (LPG) through the State Department of Culture of Bahia. The LPG was established to compensate artists for the damages caused by the Covid pandemic and allocated funds to various cultural agents and collectives across different areas such as theatre, literature, and audiovisual arts [5]. Our project was selected under the multilanguage category as a laboratory for audiovisual production for the web. With these funds, we developed a database of original silent video fragments for use in our sound improvisation practices: the project ”Cinegrafias da Costa”, which Figure 2: Live presentation of Cinegrafias da Costa at Porto Seguro’s Cultural Centre. Picure: Savi Castro Melo comprises 200 videos that can be played in random order, generating a unique audiovisual experience in each presentation (figure 2). We also acquired some audio equipment such as a 16channel sound mixer that allowed us to record all rehearsals and performances on separate audio tracks, we have compiled an extensive archive of sound recordings from the group’s work. Emphasizing the inseparable relationship between Teaching, Research, and Outreach, we are establishing a research project to edit, analyse, and organize this large audio collection, with the goal of making it publicly available as an open and free sound library. 3. REUSE OF SOUNDS This initiative arises from a need identified through our experience as a composer and educator across various curricular components in the field of Sound Creation: the scarcity of high-quality Brazilian sound libraries, especially of traditional and regional percussion instruments, available under open licenses. Most of the sounds offered on platforms such as Freesound.org are limited to genres like samba and bossa nova, which compels creators working with more experimental aesthetics or regional soundscapes to produce their own recordings. With this project, we aim to contribute to the expansion and diversification of the sonic repertoire available under Creative Commons licenses [10], by offering original and previously unreleased materials. These sounds may be used both as elements in musical compositions across different genres and in the creation of soundscapes and scores for audiovisual productions, performances, installations, and other artistic and cultural works, both within and beyond the academic sphere. Reframing the archive of material produced by the group and making it available as raw material to be used by other cultural agents is also a way of enhancing the university’s return to society, providing lasting resources that can be used by a variety of cultural practitioners. This effort supports the development of a creative network based on principles of sharing and experimentation, while also contributing to the integration of UFSB into the broader landscape of Brazilian audiovisual production. By publishing the selected collection on the Freesound.org library, we will also enable access to these sounds via the Figure 3: Original Tartyp table, by Pierre Schaeffer, source: Kelly, Ed. [9]. Playsound.org platform, previously developed by the author [15]. This platform uses sounds provided through the Freesound API [1] to be played in an open and intuitive web interface, with search functionality based on spectrogram visualization.In addition to the use of these sounds on Playsound.org, we are also developing other experimental platforms for performing with these sounds. 4. RESSOASOM RessoaSom is an experimental interface project that plays selected sounds from the Reverbera group’s archive in a randomized way, serving as a base for sound improvisations. The interface was developed with the support of two undergraduate students during the course “Digital Luthiery,” offered from March to May 2025. Our first step was to edit the recorded material in order to identify excerpts that could serve as the starting point for the creation of this sound database. With the help of undergraduate and graduate students, the selected material began to be cut, processed, and organized according to criteria based on the morphology of sound objects, as defined by Pierre Schaeffer in his Solf`ege des Objets Sonores [13], particularly in the TARTYP (Tableau R´ecapitulatif de la Typologie) shown at figure 3. His table classifies sound objects based on their properties in the time and frequency domains [12]. Chion, in his Guide to Sound Objects [2] notes that this table could encompass all possible types of sound, isolating them from their original context and “arranging them into families.” These families are organized according to morphological, temporal, and structural characteristics. While Schaeffer’s original table divides sound objects into 36 different groups, our simplified version uses only the central portion of the table, classifying the sounds into 12 groups based on duration and texture (figure 4). For temporal criteria, our sounds are categorized as: sustained, impulse, or iterative. For texture (or facture), the sounds are classified as: pitched tones, complex textures, subtle variations, and unpredictable mass variations.To streamline the group’s workflow, we standardized the duration of the sam- Figure 4: Simplified TARTYP table used for classification of the sonds sampled by the project. Figure 5: Some of the sound samples collected for the project. ples to 15 seconds, matching the duration of the films used in the Cinegrafias da Costa project. Some sounds needed to be edited to fit this duration, through trimming or extension.For the editing and analysis of the selected sounds, we employed reduced listening—that is, listening to sounds abstracted from their source or cause—as well as spectral analysis tools, which Fulop and Fitz [7]( identify as among the most useful and influential tools for analyzing acoustic signals. On figure 5 we see the waveforms over the spectrograms of some of the soudn samples collected for the project 1. As a prototype, we used a working interface provisionally named RessoaSom, which consists of a page divided into 4 columns and 3 rows. Each cell in the grid represents one of the twelve sound groups we used for classification. When a user clicks a square, a random sample from that group is played. From a programming perspective, the interface is 1Image was produced by the author using spectrogram player software developed by the first author: https:// github.com/arianestolfi/spectrogramplayer relatively simple and, at its current stage of development, requires no database or back-end processing. The working version of the tool is accessible at https://reverbera.codigo. xyz/ressoasom/ and the source code is available at https: //github.com/arianestolfi/ressoasom The development methodology followed the Lean UX approach [11], which emphasizes starting with simple, functional prototypes and gradually adding features based on user needs and technical possibilities. As a didactic tool for teaching basic web programming techniques, we began with an HTML structure containing 12 hardcoded sound objects. In the next phase, we added more sounds and developed functions to play them randomly. Later, we integrated this project with the video playback interface from the Cinegrafias da Costa project, and eventually transitioned to using the Web Audio API, enabling dynamic sample creation and real-time audio handling. At present, the platform contains 120 samples, 10 for each sound group. We plan to continue developing it as a research project, aiming to incorporate more samples and audio processing tools such as filters and compressors, as well as to expand the sound library. Our intention is to use this tool in live performances, such as the one proposed for this conference, where the interface will be used in a telematic performance in collaboration with the Female Laptop Orchestra, which will participate remotely using the Sonobus platform. The aesthetic result of this work is a tool capable of creating improvised soundscapes with a degree of randomness—by playing sounds at random—while retaining some control, as the user is able to select based on predefined morphological criteria.Throughout the research, we aimed to create a framework for organizing and classifying sound objects for realtime access in live performance settings. Building on our previous research project Experiments in Musical Interaction [14], we sought to create interfaces using technologies such as Web Audio to access and play these sounds. The Web Audio API, which enables sound processing within web interfaces, is especially valuable for developing accessible and ubiquitous tools. In this research, we have worked both on creating new interfaces for web-based performance and on further developing existing tools such as Playsound.space [15]. To ensure broader access to the sounds, we are also uploading them to Freesound.org [6] , developed by the Music Technology Group at Universitat Pompeu Fabra. Freesound is the most significant Creative Commons-based sound library today, hosting over 600,000 sounds under various types of Creative Commons licenses, which provide a legal framework for creative use and reuse according to the authors’ intentions. 5. ACKNOWLEDGMENTS Reverbera! is a project supported by the Paulo Gustavo Bahia (PGBA) public funding programs, with financial support from the Government of the State of Bahia through the State Department of Culture, under the Paulo Gustavo Law, directed by the Ministry of Culture, Federal Government of Brazil. The Paulo Gustavo Bahia (PGBA) initiative was created to implement emergency actions in support of the cultural sector, in compliance with Complementary Law No. 195, dated July 8, 2022. The project also receives support from the Federal University of Southern Bahia (UFSB). 6. REFERENCES [1] V. Akkermans, F. Font Corbera, J. Funollet, B. De Jong, G. Roma Trepat, S. Togias, and X. Serra. Freesound 2: An improved platform for sharing audio clips. 2011. [2] M. Chion. Guide to sound objects. 2009. [3] M. Chion. Audio-vision: sound on screen. Columbia University Press, 2019. [4] C. da Silveira Lima. Circula¸c˜ao audiovisual e forma¸c˜ao de p´ublico no interior da bahia: a experiˆencia do programa de extens˜ao imagina! ANAIS DE TEXTOS COMPLETOS, page 251. [5] J. R. F. de Almeida. Pol´ıticas culturais em tempos de pandemia: da lei aldir blanc `a lei paulo gustavo e suas aplica¸c˜oes no estado e munic´ıpio de s˜ao paulo. Sala Preta, 21(1):53–80, 2022. [6] F. Font, G. Roma, and X. Serra. Freesound technical demo. In Proceedings of the 21st ACM international conference on Multimedia, pages 411–412, 2013. [7] S. A. Fulop and K. Fitz. A spectrogram for the twenty-first century. Acoustics today, 2(3):26–33, 2006. [8] C. John. Silence. Lectures and Writings. Middletown, 1961. [9] E. Kelly. Deconstructing speech: New tools for speech manipulation. Organised Sound, 11:73 – 80, 04 2006. [10] L. Lessig. The creative commons. Mont. L. Rev., 65:1, 2004. [11] L. A. Liikkanen, H. Kilpi¨ o, L. Svan, and M. Hiltunen. Lean ux: the next generation of user-centered agile development? In Proceedings of the 8th nordic conference on human-computer interaction: Fun, fast, foundational, pages 1095–1100, 2014. [12] I. Neuman. Generative grammars for interactive composition based on schaeffer’s tartyp. In ICMC, 2013. [13] P. Schaeffer, G. Reibel, B. Ferreyra, H. Chiarucci, F. Bayle, A. Tanguy, J.-L. Ducarme, J.-F. Pontefract, and J. Schwarz. Solf`ege de l’objet sonore. INA GRM, 1967. [14] A. d. S. Stolfi. M´usica em rede: experimentos em intera¸c˜ao musical. 2019. [15] A. S. Stolfi, A. Milo, and M. Barthet. Playsound. space: Improvising in the browser with semantic sound objects. Journal of New Music Research, 48(4):366–384, 2019.