The objective of the present work is to detect and analyze wheezes by means of a time(flow)-frequency algorithm. An algorithm has been designed to achieve a high sensitivity to wheezing sound detection. Wheezes detection was also desired to be independent from respiratory sound power. Both objectives have been achieved. Automatic measurements have been compared to the clinical auscultation for the forced exhalation segment between 1.2 and 0 1/s. Detection algorithm validation has been done in collaboration with a medical doctor. A good sensitivity (100% to 71% of wheezing segments as a function of flow level) and also a very good specificity for non-wheezing episodes (100% for high and medium flow segments and 88.2% for low flow segment) has been shown by the algorithm. Objective time(flow)-frequency wheezes parameter extraction has been done for the flow range from 1.2 to 0.2 l/s, during forced exhalation. Wheezes have been detected in both analyzed groups: asthmatics (N=16) and control subjects (N=15), as reported in other articles and traditional auscultation works. Significant differences between two groups have been found for the mean number of wheezes detected in a patient maneuver (p=0.0003). A higher significance than in other works has been observed for this parameter. For frequency and other analyzed parameters differences were also significant (0.0112>p>0.0307).