Abstract: The application of deep learning in chemical metrology has been receiving increasing attention，particularly in the field of spectroscopy.Spectral analysis technology，as an important component of chemometrics，has long played a key role in molecular analysis，structural identification，and quantitative measurement.With the continuous development of deep learning algorithms and computational capabilities，the challenges faced by traditional spectral data analysis methods are gradually being effectively addressed.This research reviews the current applications，technological advantages，and future trends of deep learning in the four major spectroscopic techniques—Infrared spectroscopy (IR)，Ultraviolet spectrum (UV)，Nuclear Magnetic Resonance (NMR)，and Mass Spectrometry (MS).It analyzes how deep learning，through big data training，automated feature extraction，and nonlinear modeling，enhances the accuracy and efficiency of spectral analysis.With the improvement of computational power，further optimization of deep learning models，and the continuous accumulation of big data，the application of deep learning in chemical metrology，particularly in the field of spectral analysis，is advancing towards a more intelligent and automated direction.This is bringing about revolutionary changes in chemical analysis technology and providing strong support for scientific research and industrial applications.

Key words: Deep Learning（DL）, Stoichiometry, Infrared spectroscopy（IR）, Ultraviolet spectrum（UV）, Nuclear Magnetic Resonance（NMR）, Mass Spectrometry（MS）