It is curious that whether the typical herringbone pattern of bird flight feathers plays a role in attenuating aeroacoustic noise. Being motivated by this, experiments are conducted to investigate noise reduction of the NACA0012 (National Advisory Committee for Aeronautics) airfoil-based model with one side surface covered by bio-inspired herringbone patterns of riblets. The herringbone-ribbed surface is defined by the divergent angle β (= 60°) of the riblets pattern, the spanwise wavelength λ (= 0.2c and 0.4c) of the pattern, and the riblet height h (= 0.6%c and 1.2%c), where c is the chord length of the airfoil. While far-field sound pressure fluctuations are measured via microphones in an anechoic wind tunnel, flow fields around the model are captured using particle image velocimetry (PIV) in a water tunnel. The effective angle of attack of the test models ranges from αeff = −11.1° to 11.1° and Reynolds number considered is from Re = 2.3 × 105 to 7.8 × 105. Compared with the baseline smooth models, the models with the riblet pattern on the pressure side are able to substantially suppress the tonal noise, associated with considerable reduction in the overall noise. The reduction in the overall sound pressure levels of the tonal noise and the overall noise are up to 21.3 dB and 20.5 dB, respectively, at αeff = −2.2° and Re = 3.6 × 105. The noise reduction is attributed to the transition of laminar to turbulent boundary layers over the herringbone-ribbed surface, particularly in the saddle location where the spanwise-repeated herringbone pattern converges. Behavior of shear layers separating from the trailing edge of the model is examined, corroborating the proposition that the acoustic feedback loop is impaired by the herringbone-ribbed surface.