M(OTHER)
Reconstructing Maternal Voice Through Bio-sonic AI and Breast Milk Forensics
2022-2025
M(Other) is a spatial bio-sonic installation that reconstructs human voices from the breast milk of wet-nurse donors, transforming molecular data into an immersive sound experience. Using DNA profiling, genetic material is sequenced, decoded, and reassembled through AI-driven synthesis to generate three distinct lullabies.
The project is rooted in the complexity of the voice genome—an intricate biological code woven into DNA yet difficult to fully interpret. By encoding voice at the molecular level, M(Other) collapses the boundary between inheritance and intervention, asking how identity can be reclaimed when the most intimate aspects of life can be programmed.
This complexity echoes the history of wet nurses, whose reproductive labor was vital and coerced—indispensable yet unacknowledged. M(Other) embodies this dual erasure by surfacing both inherited and synthetic voices, confronting questions of authorship, memory, and maternal identity.
Voice and Vowel Acoustics Analysis:
- Speech Data Collection: Speech recordings from isolated vowels, sustained sounds, single words, and read passages provided diverse acoustic measures.
- Key Metrics: Voice pitch (Fo), pitch variability, vowel formants (F1 to F4), and metrics like VSA4 and FCR to assess clarity and vocal tract length.
- Lifespan Analysis: The data revealed variations in pitch and formant values across different ages and genders, offering insights into vocal evolution.
Voice Analysis: AI tools analyzed DNA-derived vocal markers to reconstruct vocal tones with high precision.
Spectrograms of three DNA-driven lullabies from M(Other)
These visualizations capture the spectral energy of three AI-generated lullabies based on breast milk-derived DNA. Each work translates bio-sonic identity into affective and rhythmic vocal patterns, demonstrating how maternal biomatter can be reconfigured into sonic presence.
Reclaiming breast milk as epistemic material, the project presents it as a site of emotional, genetic, and political force. M(Other) challenges dominant systems of knowledge by rewriting how we hear care, archive absence, and imagine maternal presence in the age of synthetic reconstruction. It breaks the wall between flesh and data, between what we inherit—and what we choose to remember.
In this space, M(OTHER) does not simply archive lost voices; it confronts the instability of voice itself, asking whether technology reclaims presence or manufactures absence. Visitors stepping into M(OTHER) will find themselves in an intimate sound-proof chamber where echoes of voices intertwine with the faint scent of newborn skin, evoking first tender embraces. As they move, they activate lullabies linked to different genetic profiles, immersing them in an experience where memory, sound, and biological data merge. The walls, reminiscent of padded rooms once used for silence and control, transform into a site of resistance, where voice—once suppressed—becomes the central force.
