Skip to main content
SearchLoginLogin or Signup

The Historically Informed AI: Johann Sperger through time and bass

Johann Sperger's sonata for contrabass and viola revisited through AI generated soundscapes

Published onAug 03, 2023
The Historically Informed AI: Johann Sperger through time and bass


Johann Sperger (1750-1812) was a composer and contrabass virtuoso. Even today, Sperger holds the record for composing the largest number of contrabass concertos. In his practice, Sperger thoroughly explored the technical and expressive boundaries of his instrument in a dynamic process of reciprocal negotiation between the performer’s and instrument’s agencies. At the time, Sperger’s explorations of the highest register of the contrabass posed a wild challenge to its teleological identity. In this work, we aim to further reflect upon his work by combining historically informed performance and musical instruments (contrabass and viola) with contemporary forms of notation and digital luthiery, including graphic scores, specifically designed controllers and neural synthesis. In order to join our apparently distant temporal and musical dimensions, we retraced the route from Trieste to Venice that Sperger traveled while revising his Sonata for Contrabass and Viola (1789). We collected sound samples that will be used to train our neural synthesis model. Sensors attached to the performers’ arms will render their movements as the manifestation of the composer’s, the performers’ and the instruments’ interactions. These movements will be used in the decoding module to inform the navigation of the model's latent spaces. The sound output of the decoder will be fed to the encoding module, generating a graphic score which will in turn influence the execution.


In a bid to create a representation of Sperger’s biography that was not bound to the text/book format, we set Trieste and Venice (the two locations mentioned in the autograph manuscript of Sperger’s 1789 sonata for contrabass and viola) as the geographic parameters of what would become a multimedia exploration of Sperger’s life and music. “Trieste” appears at the start of the manuscript score, indicating that the sonata was completed there. “In Venezia” is written at the end of the manuscript parts, indicating that the parts were copied on the way to and completed in Venice (or copied in their entirety in Venice).

In 1789, Trieste and neighbouring Venice were two competing port-cities on the Adriatic coast. The former, a cosmopolitan city from the Habsburg Austrian Empire and the latter, the main city of the decadent Republic of Venice. It is most likely that Sperger travelled by sea since there was a direct connection between the two port-cities, and the route by land left travellers at the mercy of briganti, the local bandits. Unfazed by the threat of highway robbery, and in lieu of a direct modern-day route from Trieste to Venice by boat, we opted for a land-based pilgrimage in June of 2022.

In this performance we therefore combine historically informed performance and historical musicology, in-loco collection of audio material during our journey to the two port-cities with contemporary forms of notation and digital luthiery, including graphic scores and neural synthesis.

Historically informed performance (HIP) is a philological approach of so-called classical music that attempts to recreate the ways music was conceived and performed. It considers both the technology — historical instruments — as well as the technical and stylistic practices, obtained through an hermeneutical approach to texts and iconography. This project can be framed within the post-HIP movement as it incorporates contemporary technology[1]. Within this movement, contemporary composition using historical instruments is widespread and can also include the use of historical style and forms, recomposing sections of existing historical works, and augmenting them with electronics sounds. Examples include Mark Grey’s electronic composition and sound design for the recitatives in the 2017 rendering of Handel’s Aci, Galatea e Polifemo directed by Christopher Alden[2], Patricia Alessandrini and Freida Abtan’s Orpheus Machines (2015)[3], and new compositions for baroque instruments and electronics commissioned by The Nordic Affect.[4] Our intention is to build on the precedent set by Sperger himself with a performance at the forefront of the post-HIP interdisciplinary artistic research. To continue pushing Sperger’s exploration on the perceived limits of the instruments and their expressive potential we will instantiate a dialogue between Sperger’s music, the musicians’ bodies and a neural synthesis model, mediated by graphic notation.

An initial work combining the audiovisual material with a performance of the second movement of Sperger’s sonata can be viewed here. In the video provided, the movement is played as written, thus the performance and the soundscapes seem to follow independent parallel paths, with the latter interpreting the emotional narrative of the former. We believe that the final output of our research should be performative rather than descriptive, as Sperger’s music (and this movement in particular) captures an immanent and dynamic process of negotiation with the instrument’s technical boundaries. In the proposed performance we aim to extemporarily interweave the audiovisual material and live performance, with substantial formal changes and freedom of interpretation, in order to open a portal between the aesthetics of past and present times by extending the expressive boundaries of the contrabass and the viola.

Instead of composing the soundscape as in the attached video, we will curate a dataset that will train a RAVE neural synthesis model [5]. The dataset will consist of the sound samples we collected in our journey between Trieste and Venice. The movements of the performers interacting with their instruments will be captured by sensors and draw a graphic score in IanniX in real time [6]. The slightly deferred playback of the score, visible to the audience through moving cursors, will control the navigation of the model’s latent spaces. How this will in turn influence the musician’s performance will be subject to experimentation in the forthcoming months. In our practice-based research we will explore extemporary ornamentation, movement as notation, dynamic scores, and the use of controllers to mediate between the instrument and the score. In order to discover alternative ways of engaging with the performative space, we will work with the following research questions:

  • how does musical practice change with the advent of artificially intelligent technologies?

  • how music may be composed and represented in order to facilitate dialogue?

  • How to re-contextualise historically informed practice in order to interpret the tension towards research of past composers?


Darija Andelic-Andzakovic

Darija Andzakovic (Germany/Austria) is a bassist and musicologist, specializing in historic bass instruments and music of the late-18th century. She is currently a doctoral candidate at the University of Music and Performing Arts in Vienna and has worked with Les Arts Florissants, Musique des Lumières, Neues Orchester Basel and Bach Collegium Zurich. 

Natalia Duarte Jeremias

Costa Rican violist specialized in historical and modern viola. She has a BA in anthropology and music, and master’s degrees in modern viola (U. of Arizona), early music (ESMUC) and instrumental pedagogy (Conservatory of Padova). She collaborates with several historically informed orchestras and chamber ensembles such as Innsbruck Festival Orchestra, Ensemble laBarocca di Milano, Academia Montis Regalis and others. As a music teacher, she is interested in music improvisation and using music in participatory processes.

Nicola Privato

I’m a PhD candidate in Cultural Studies, conducting my research at the Intelligent Instruments Lab. Previously, I studied Electronic Music at the Conservatory of Padua (MA), Jazz Improvisation and Composition at the Conservatory of Trieste (BA) and Modern Languages and Cultures at the University of Padua (BA). In the last ten years I have been curating musical events and festivals, composing, performing and teaching music. My current interests include alternative forms of notation, improvisation, composition, and Human-Computer Interaction in performative contexts. My project focuses on AI explainability in music performances. My works and performances can be viewed here.

Technical Rider

We will provide:

  • Instruments

  • Computer and sensors

  • Router for ad-hoc network

  • Audio Interface

We require:

  • Music Stands

  • Projector

  • Audio System (Stereo or Multi-channel)

  • Two microphones

Setup Details

We will need to project the score for the audience, but also to the musicians. A dedicated screen for this would be ideal.

A table will be necessary, and cables from the audio interface to the audio system. We will need to power: laptop, audio interface, portable router.

Depending on the dimensions of the room

Individual works and performances of each member can be viewed at the following links:

We believe that the sound focused concert will be best for this performance (31st August).

Programme Notes

Johann Sperger (1750-1812) was a composer and contrabass virtuoso that thoroughly explored the technical and expressive boundaries of the instrument. Sperger worked on his Sonata for Contrabass and Viola (1789) while travelling from Trieste to Venice, as he states in the manuscript. For this work, we traced his route and recorded the soundscapes we encountered. We used these sounds, as well as our performative gestures, to train and navigate a RAVE model, with the aim of exploring, through a historically-informed approach, the relationship between past and present technologies, virtuosity, compositional methodologies and performative practices.

No comments here
Why not start the discussion?