That Stinks! New GenAI Multimodal type: Smell to Text

The September 1 issue of Science has a great article highlighting a groundbreaking AI program, specifically a graph neural network, that excels at emulating human olfactory abilities, especially with basic scents.

The study results, reported show that the program, a so-called graph neural network, is excellent at imitating human sniffers, at least when it comes to simple odors. It reliably predicted what the volunteers smelled, a feat sensory biologists have been working toward for decades. It also predicted the smells of 500,000 other molecules, with no need to make or sniff them.

Olfaction is the sensation of smell that results from the detection of odorous substances aerosolized in the environment. Along with vision, taste, hearing and balance, olfaction is a special sense. Humans detect odors through the olfactory epithelium of the nose, whereas most insects detect odors through their antennae.

The system not only accurately identified the scents detected by human participants—a milestone that sensory biologists have aimed to achieve for years—but also predicted the aromas of half a million other compounds without the need to produce or smell them firsthand.

This innovative method holds vast promise in accelerating the quest for enhanced aromatic consumer goods.

Companies using AI for digital olfaction

Aryballe, is a startup that uses artificial intelligence and digital olfaction technology to mimic the human sense of smell. The company helps their business customers turn odor data into actionable information.

Companies like Aryballe will potentially be collaborating on projects that will create digital odor libraries for companies, or even creating devices that help COVID-19 patients recover their sense of smell.

Practical Applications for Digital Olfaction inBusiness

● Craft the quintessential “new car” aroma that many of us love in the auto sector.

● Foresee when machinery in industrial or automotive contexts might need a touch-up.

● Intuitively sense when food starts to go bad within consumer electronics.

● Make informed decisions about the acceptance or rejection of raw materials.

● Expedite the R&D process when innovating new flavors and drinks.

● Guarantee that the scent of personal care items, such as deodorants and shampoos, remains consistent over prolonged periods.

● Enhance the commuter experience in public transport with the release of soothing fragrances.

● Develop cutting-edge personal care gadgets and health monitors that leverage scents to identify potential problems and promptly notify users.

● Environmental Monitoring: Detect harmful pollutants or chemicals in the air, especially in industrial zones or polluted cities, and send alerts for necessary interventions.

● Agriculture: Monitor soil health or detect pests based on their specific odors, leading to more targeted and effective farming interventions.

● Retail Experience: Stores can change ambient smells based on the time of day or customer preferences, enhancing the overall shopping experience.

● Fire Detection: Before smoke becomes visible, some fires emit distinct odors. Digital olfaction can serve as an early warning system in such scenarios.

● Travel & Tourism: Hotels or tourist spots can offer signature scents to make the visitor’s experience memorable. This can also help in branding and customer loyalty.

● Fitness & Well-being: Monitor body odor to provide insights into personal health metrics, hydration levels, or even the need to increase or decrease workout intensities.

● Forensics & Security: Use digital olfaction devices at crime scenes to detect traces of specific chemicals, substances, or even human presence, aiding in investigations.

With such a wide array of potential applications, digital olfaction is poised to redefine many sectors, making them more efficient, user-friendly, and innovative.

David Cronshaw