Generative AI masters the art of scent creation
A novel model using generative diffusion neural network automatically creates custom fragrance recipes, based on user input of scent descriptors
2025-04-23
(Press-News.org)
Addressing the challenges of fragrance design, researchers at Institute of Science Tokyo (Science Tokyo) have developed an AI model that can automate the creation of new fragrances based on user-defined scent descriptors. The model uses mass spectrometry profiles of essential oils and corresponding odor descriptors to generate essential oil blends for new scents. This breakthrough is a game-changer for the fragrance industry, moving beyond trial-and-error, enabling rapid and scalable fragrance production.
Designing new fragrances is crucial in industries like perfumery, food, and home products, where scent significantly influences the overall experience of these products. However, traditional fragrance creation can be time-consuming and often depends on the skill and expertise of specialized perfumers. The process is typically challenging and labor-intensive, requiring numerous trial-and-error attempts to achieve the desired scent.
To automate this process, a research team, led by Professor Takamichi Nakamoto from Institute of Science Tokyo (Science Tokyo), developed an AI model called Odor Generative Diffusion (OGDiffusion). This model utilizes generative diffusion networks, a type of machine learning model that learns to create new content by reversing a noise process informed by existing data. These models are already widely employed to generate images and text, and the team has adapted this technology to create new fragrances. Their findings were published in IEEE Access on March 27, 2025.
The system operates by analyzing the chemical profiles (mass spectrometry data) of 166 essential oils, which are labeled with nine odor descriptors (such as “citrus” or “woody”). When users specify desired scent characteristics, AI generates a corresponding chemical profile (mass spectrum) that aligns with those descriptors. It then calculates the mix of essential oils needed to recreate that scent using a mathematical method called non-negative least squares.
“Our diffusion network uses patterns in mass spectrometry data of essential oils to generate new fragrance profiles in a fully automated, streamlined, and data-driven approach while maintaining high-quality data output. By eliminating human intervention and molecular synthesis from the process, we provide a fast, general, and efficient method for fragrance generation,” explains Nakamoto.
While existing AI-based fragrance generation models, they have been developed, they rely on proprietary datasets and still require expert input. The primary advantage of the new method is its ability to automate the creation of new scents completely. Moreover, as the system produces fragrances based on essential oil recipes, the final scent can be easily recreated.
Further, the team conducted human sensory tests to evaluate whether the AI-generated fragrances align with the intended scent profiles. In a double-blind setup, 14 participants were tasked with matching AI-generated fragrances to appropriate descriptors (such as “citrusy” or “floral”). Participants were consistently able to identify the correct fragrance, demonstrating that the system could produce scents that met people's expectations. In another test, participants distinguished between two scents: one designed to express an additional specific odor descriptor and original scent without that descriptor. They reliably selected the scent that matched the target descriptor, indicating that the model generates clear and identifiable scent profiles.
Nakamoto’s model—the first of its kind—heralds a future in which AI transforms scent design. “This approach represents a significant advancement in aroma design,” states Nakamoto. Adding further, he says, “By automating the generation of mass spectra corresponding to desired odor profiles, the OGDiffusion network offers a more efficient and scalable method for fragrance creation. Moreover, even a novice can create an intended scent to make scented digital contents”
In summary, this innovative method allows for faster and more flexible scent design, with potential applications across various industries. By leveraging AI for scent generation, the OGDiffusion model demonstrates that computers can indeed possess a nose for creativity.
Nakamoto Lab
http://silvia.mn.ee.titech.ac.jp/html_en/index_en.html
About Institute of Science Tokyo (Science Tokyo)
Institute of Science Tokyo (Science Tokyo) was established on October 1, 2024, following the merger between Tokyo Medical and Dental University (TMDU) and Tokyo Institute of Technology (Tokyo Tech), with the mission of “Advancing science and human well-being to create value for and with society.”
https://www.isct.ac.jp/en
END
ELSE PRESS RELEASES FROM THIS DATE:
2025-04-23
People with Alzheimer’s disease may retain their ability to empathise, despite declines in other social abilities, finds a new study led by University College London (UCL) researchers.
The researchers found that people with Alzheimer’s disease scored slightly higher on a measure of empathy than peers of the same age with mild cognitive impairment, despite scoring worse on other measures of social cognition such as recognising facial emotions and understanding the thoughts of others.
The authors of the study, published in Alzheimer’s & Dementia ...
2025-04-23
Primal world beliefs (“primals”) capture understanding of general characteristics of the world, such as whether the world is “Good,” “Safe,” and “Enticing.” In a new study, researchers analyzed responses from children, mothers, fathers and then later, young adults in Colombia, Italy, Jordan, Kenya, Philippines, Sweden, Thailand, and the United States to learn about their world beliefs.
Through a larger project called Parenting Across Cultures, children ages 8-16 years old (50% female) and their mothers and fathers, reported responses about neighborhood danger, socioeconomic status, parental ...
2025-04-23
Research published in the Journal of Management Studies provides new insights on how parental leave policies, involving their time and money components, might affect whether mothers start a new business.
Contrary to the common belief that generous parental leave discourages entrepreneurship, the study finds that offering more time off through parental leave can promote mothers’ entrepreneurship.
In two complementary studies, the researchers found that the duration of parental leave—yet not the amount of money paid—seems to shape mothers’ entrepreneurial activity. ...
2025-04-23
A study published in Aging Cell has identified immune resilience as a key driver of salutogenesis—the active process of promoting health and well-being.
By analyzing data from 17,500 individuals across various life stages, investigators uncovered the importance of immune resilience involving TCF7, a gene essential for maintaining immune cell regenerative potential, in fostering healthy aging and longevity.
The research indicates that immune resilience counters three major factors of aging and mortality: chronic inflammation, immune ...
2025-04-23
Hip replacement surgery, or total hip arthroplasty (THA), can lessen pain and improve function in individuals with hip osteoarthritis, but some patients continue to experience long-term physical deficits—including muscle weakness, decreased functional mobility, and increased fall risk—after the procedure. New research published in the Journal of Orthopaedic Research reveals that a patient’s muscle quality before THA may predict their risk of such suboptimal recovery after surgery.
In the study, 10 people undergoing THA underwent imaging tests before surgery. Patients whose imaging results indicated poor muscle quality were more likely to ...
2025-04-23
Results from a study published in Brain and Behavior suggest that traumatic events may contribute to the development of eating disorders among Palestinians. Also, certain sociodemographic factors—such as educational level, gender, region, and age—were linked to a greater likelihood of experiencing eating disorders.
In the study of 580 adults, including 320 males and 260 females, who were recruited online, individuals who reported living through traumatic events were more likely to experience restraint eating, eating concern, weight ...
2025-04-23
New research published in BJOG: An International Journal of Obstetrics & Gynaecology found that mothers who are anemic in early pregnancy face a higher likelihood of giving birth to a child with a heart defect.
The study assessed the health records of 2,776 women with a child diagnosed with congenital heart disease who were matched to 13,880 women whose children did not have this condition.
Investigators found that 4.4% of children with congenital heart disease and 2.8% of children with normal heart function had anemia. After adjusting for potential influencing factors, the odds of giving birth to a child with congenital heart disease was 47% higher among anemic mothers.
“We ...
2025-04-23
The back-and-forth shift in climate policy between the Biden and Trump administrations has created uncertainty about future directions related to addressing climate change. A new study in International Studies of Economics examined the impact of climate policy uncertainty on world energy stock returns.
The study found that a rise in climate policy uncertainty causes stocks to plummet in individual countries, regions, and the world energy stock markets, as investors perceive that climate policy uncertainty could ...
2025-04-23
Multiple climate “tipping points” are likely to be triggered if global policies stay on their current course, new research shows.
Scientists assessed the risk of “tipping” in 16 different parts of the Earth system – ranging from the collapse of major ice sheets to the dieback of tropical coral reefs and vast forests.
Based on current policies and the resulting global warming, their most conservative estimate is a 62% risk of triggering these tipping points on average.
However, more sustainable future pathways – with lower greenhouse gas emissions – significantly ...
2025-04-23
Colm Nestor, senior associate professor in medical genetics, has been awarded the 2025 Onkel Adam Prize for outstanding research at the Faculty of Medicine and Health Sciences at Linköping University, Sweden. One of his research areas is gender differences in susceptibility to autoimmune diseases and infections.
Having studied both plant biology and programming, Colm Nestor happened to have the right skills at the right time. He had just finished his undergraduate programme at the turn of the millennium when the human genome was mapped through ...
LAST 30 PRESS RELEASES:
[Press-News.org] Generative AI masters the art of scent creation
A novel model using generative diffusion neural network automatically creates custom fragrance recipes, based on user input of scent descriptors
