Unmixing the World: Making Sound as Programmable as Code | Jessica Powell, Audioshake @scaleconference
Unmixing the World: Making Sound as Programmable as Code | Jessica Powell, Audioshake  @scaleconference
Uploaded October 2025 | Updated November 2026, 2 days ago
Sound is one of the world’s richest but least accessible data sources. Most audio exists as a single, tangled mix—impossible to edit for humans and noisy for machines to understand. This talk explores how advances in audio separation are unmixing the world: splitting sound into its component parts so it can be edited, searched, and programmed like code.

For humans, this means new creative control, immersive experiences, and copyright-safe workflows. For machines, it provides the fine-grained data needed to power multimodal training, boost speech recognition, and enable entirely new AI applications. By transforming raw sound into structured, usable data, separation is becoming the intelligence layer for audio—laying the foundation for the next generation of media, AI, and human-computer interaction.

Jessica Powell, the CEO and co-founder of Audio Shake, delivered a presentation on her company's audio source separation technology. [00:52]

She began by outlining the shared problem both humans and machines face in understanding and filtering complex, real-world sound. [01:33]

Powell explained that Audio Shake's technology addressed this by splitting audio into its component parts, such as separating a voice from noise or one instrument from another. She detailed the technical challenges of balancing processing speed with quality and asserted that Audio Shake outperformed open-source models on both fronts. [04:34]

The presentation then covered a wide range of applications, first for human-centric creative workflows in music, film, and broadcasting, and then for machine-centric uses like improving voice AI accuracy, enabling the editing of generative media, and creating higher-quality training data for multi-modal AI models.
Unmixing the World: Making Sound as Programmable as Code | Jessica Powell, AudioshakeGet an exclusive look inside Metas hardware lab!CEO and Co-Founder of Audioshake.ai, Jessica Powell | Episode 4 @Scale PodcastScaling Privacy Infrastructure for GenAI Product Innovation | Ming Qiao and Ram Rathnam, MetaUltra Low Latency Connect | Ishan Khot and Hani Atassi, MetaAI is changing fast!Keynote | Ime ArchibongWhat Infrastructure Will Power the AI-Native Future?What does it take to scale AI infrastructure at unprecedented speed?AI is reshaping how we build, scale, and innovate — Register for AI & Data today!Learn how to scale at @Scale!What does it take to build products that evolve and grow?
@Scale |

Unmixing the World: Making Sound as Programmable as Code | Jessica Powell, Audioshake

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER