Many noise reduction algorithm improves overall speech quality but little progress has been made to improve speech intelligibility. The paper proposed by Gibak Kim and Philipos C. Loizou uses new binary mask approach to improve speech intelligibility, for our paper we have used the same approach but used parametric Wiener filter algorithm for the gain function computation. Subjective and objective measures tests were conducted to evaluate the overall speech enhancement quality and speech intelligibility. For the objective measure, the segmental signal-to-noise ratio parameter was calculated. The results indicate improvement in segmental signal-to-noise ratio for speech corrupted by noise at 0dB and -5 dB SNR levels for Helicopter noise and Car noise.