In this paper, a fast and quick modulation recogniser for different types of analog modulations is developed. This recogniser utilises the decision-theoretic approach in which the extracted key features are compared with suitable thresholds to decide about the modulation type. In this recogniser, all the key features used are derived from the complex envelope of a signal. Thus, this recogniser is fast compared with those presented in the available references. Computer simulations for different types of band-limited analog modulated signals corrupted with band-limited Gaussian noise sequence have been carried out. It is found that this recogniser performs %veil and all the analog modulation types of interest - AM, DSB, LS13, US13, and FM - have been classified with success rate 100 % a't the SNR of 10 dB.