a.s.ist Launches 7-Day Free Trial for "AutoStatSpectra," Automated Spectral Analysis Software Using Bayesian Statistics

a.s.ist (株式会社a.s.ist) is promoting DX in the manufacturing industry with white-box AI based on research knowledge cultivated at the University of Tokyo.

a.s.ist will start a 7-day free trial for its spectral analysis software "AutoStatSpectra," which is based on Bayesian statistics. All functions will be available without restrictions during the trial period. *The claim "world's first" in the title is based on the company's own research as of April 2026.

**Development Background:** In spectral analysis methods like XPS (X-ray photoelectron spectroscopy), FT-IR (Fourier transform infrared spectroscopy), and NMR (nuclear magnetic resonance), the separation of overlapping peaks and the setting of peak numbers, parameters, and function forms significantly influence analysis results.

However, conventional curve fitting software often relies on the analyst's experience and trial-and-error for these settings, leading to issues in objectivity and reproducibility of results.

AutoStatSpectra achieves automatic estimation of peak numbers, peak parameters, and peak function forms, along with quantification of estimation uncertainty, through Bayesian inference. It provides statistically supported spectral analysis that is independent of the analyst's subjectivity.

**Key Features of AutoStatSpectra:**

**Automatic Estimation of Peak Number, Parameters, and Function Forms:** Compares models within a specified range and automatically selects the optimal number of peaks based on posterior probability. It also automatically estimates parameters like peak height, center position, and half-width, as well as peak function forms (Gaussian, Lorentzian, pseudo-Voigt) suitable for the data. The quantitative evaluation of uncertainty for each estimated value enables highly reproducible peak fitting that is less dependent on the analyst's experience and trial-and-error.

**Diverse Background Correction:** Supports multiple background correction methods, including constant background, linear background, and Shirley method for XPS, allowing flexible selection of preprocessing methods according to various measurement techniques.

**Batch Analysis of Multiple Files (Multi-file Workspace):** Multiple spectral data can be grouped into one workspace and processed in batches under common analysis conditions. This supports efficient analysis of large datasets for condition studies and time-course comparisons.

**Automatic Report Generation:** Analysis results are automatically output as PDF/Word reports. Comprehensive reports including fitting plots, model comparison tables, parameter lists, and posterior distribution plots significantly shorten the workflow from analysis to reporting. Data export to CSV and Excel is also supported.

**Supports Various Data Formats:** Widely compatible with various data formats such as CSV, Excel, VAMAS, and SOF-H.

**About a.s.ist Co., Ltd.:** a.s.ist is a startup utilizing advanced data analysis technologies such as white-box AI, Bayesian inference, and sparse modeling, based on research insights cultivated at the University of Tokyo.

With strengths in data analysis for materials development and physical property experiments, the company supports problem-solving in research and manufacturing sites through the development and sales of automated analysis software and the provision of custom DX solutions. Furthermore, it is also engaged in the utilization of new AI technologies such as Physical AI.

Company Name: a.s.ist Co., Ltd. Representative Director and President: Yui Hayashi Location: 5-17-24 Iko, Adachi-ku, Tokyo Established: June 2021 URL: https://www.a-s-ist.com