Speech intelligibility tests are conducted by the pronunciation of a list of monosyllabic words. In this case in the Malay language and a separate and parallel test with English language. Generally the results of such a test are more pessimistic than that in the comprehension of a discourse or speech. In the former the words have no relation to any other word on the list and are by definition a random sample. On the other hand a speech has interconnected ideas and provides a coherent synopsis of concepts. The audience can guess therefore the general outline of the speaker's message.

Speech intelligibility tests should also be conducted by a masking noise overlaid on to the test syllables to simulate any intrusive background influence, which is typical of a situation whether it be ventilating or air conditioning systems. The masking must be controlled at a specific increasing dB level and be set at a range of frequencies typical of that of the background noise. It is of course essential to conduct this test with minimum background at the start. The listening audience should be dispersed throughout the hall and write down the syllable which they think is being spoken.