| 1. |
- Boersen, Ronald, et al.
(författare)
-
Chatterbox: an interactive system of gibberish agents
- 2020
-
Ingår i: Proceedings of 26th International Symposium of Electronic Arts (ISEA 2020). - 9782981641328
-
Konferensbidrag (refereegranskat)abstract
- We present the interactive multi-agent system Chatterbox, as part of the sound art installation Translanguaging, exploring the notion of translanguaging as a mediation of multilingual and intercultural communication. We discuss the act of languaging as a dual process comprising both semantic language communication, as well as paralanguage that relates to the affective, personal, and cultural aspects related to translanguaging. Through the creation of the Chatterbox agent, generating gibberish vocal streams devoid of semantic content, we aim at highlighting the paralinguistic dimension of languaging. The agent model comprises a kind of gradient map, clustering a segmented corpus of vocal sounds in the latent space of a self-organized map, according to its paralinguistic fingerprint. We utilize Factor Oracles for the creative generation of novel utterances of paralanguaging gibberish by the agent. Incorporating simple subsumption architecture inspired rules, we further moderate the interaction between the gibberish agents, creating rich and complex multi-agent behavior in “paralanguaging discussion”. We outline the artistic and technical considerations in developing our Chatterbox agent throughout the paper. We share several observations made throughout the process of creating the Chatterbox agent, highlighting some of the connections between the notion of (trans)languaging and the implementation of our model.
|
|
| 2. |
- Tatar, Kivanc, 1988, et al.
(författare)
-
Latent Timbre Synthesis: Audio-based Variational Auto-Encoders for Music Composition Applications
- 2020
-
Ingår i: Neural Computing and Applications. - : Springer Science and Business Media LLC. - 0941-0643 .- 1433-3058. ; 33:The Special Issue of Neural Computing and Applications: “Networks in Art, Sound and Design.”, s. 67-84
-
Tidskriftsartikel (refereegranskat)abstract
- We present the Latent Timbre Synthesis, a new audio synthesis method using deep learning. The synthesis method allows composers and sound designers to interpolate and extrapolate between the timbre of multiple sounds using the latent space of audio frames. We provide the details of two Variational Autoencoder architectures for the Latent Timbre Synthesis and compare their advantages and drawbacks. The implementation includes a fully working application with a graphical user interface, called interpolate_two, which enables practitioners to generate timbres between two audio excerpts of their selection using interpolation and extrapolation in the latent space of audio frames. Our implementation is open source, and we aim to improve the accessibility of this technology by providing a guide for users with any technical background. Our study includes a qualitative analysis where nine composers evaluated the Latent Timbre Synthesis and the interpolate_two application within their practices. 2.14.0.0
|
|
