Estimation of volume fractions in multiphase flow is typically a major concern in process sectors especially the oil industries. The aim of this study was to develop an online technique to address the demands of monitoring the volume fractions of the most challenging multiphase fluid, oil/ gas, flowing in a pipeline based on process sampling and acoustic chemometrics. An experimental rig, on which acoustic sensor and a newly constructed sampling device were mounted, was utilized for this investigation. The sampler was mounted on a bypass loop in the rig and reference samples of the fluid were collected automatically and evaluated. The volume fractions of each sample were compared with a priori known input volumes. It was concluded that the sampling error was quite low. Also, the acoustic signals of various mixtures of oil/gas flowing through the rig were recorded. Thereafter, multivariate calibration (Partial Least Square-Regression, PLS-R) method, based on test set validation, was utilized for the prediction of various volumes of oil, as well as gas in the oil/gas mixtures flowing in the rig. The pressure drop of the system was 2.5 bar and a constant temperature was maintained throughout the experiment. This study does not cover temperature variations and static pressure variations. The experimental results from the feasibility study were promising, but more experimental work is needed to use this technique for online monitoring in the oil industry.