ICMC/25/Installation/Screening/06 ludus vocalis
ludus vocalis
- 21:41 extremely fast, scattered chunks of videos and audio recordings of various sounds that come from the mouth. Germs, Laughs.
- It's very effervescent. Almost hard to take in, like pop rocks in soda inside the nose while going down a waterslide.
- the Sound/Speaker/Multichannel Sound/Spatial explosion is gross and delightful. Any kid would love this. At the same time, it's not trying to shock in such a way that wouldn't be appropriate for a child.
- There is an orgamic one, FYI parents ... but it's not graphic. It's kind of "fireworksy" and "auto-Tuney" - it's hard not to smile. The graphics here are not as literal as some. many times, it's graphical in a geometric way.
- The contrast between the different sections in terms of sonority, visual appearance, texture, etc. it's very #Creative and well-done.
#Description
- ludus vocalis (2024)
- 25min work for Sound/Speaker/Multichannel/8 Channel audio, Video/4k
- explores, reimagines paralinguistic vocal sounds as musical objects
- laughing
- crying
- screaming
- gasping
- moaning
- ... sounds carry emotional and largely cross-cultural qualities
- sixteen short vignettes
- each focusing on a specific type of nonverbal vocalization
Source Material
- manually curated and gathered from various online platforms
- Sound/Epidemic (https://www.epidemicsound.com/)
- Splice (https://splice.com/)
- FreeSound (https://freesound.org/)
- samples—586 audio files in WAV format
- (~40min)
- organized into distinct paralinguistic categories
- smaller, focused audio corpora created
- Each corpus employed different segmentation strategies
- and audio analyses
- to facilitate deeper exploration of the source material.
Video
- developed using AI-generated video, primarily created via AI/Text/To/Video text-prompts with RunwayML (https://runwayml.com/), as raw material.
- Broadly speaking
- generated videos were intended to represent mouths enacting the different kinds of vocalizations that each movement was based on
- and the “failure” of the generative model to adhere to certain prompts was often embraced.