The rhythm measures are one of the primary means of differentiating languages and their dialects. They are broadly classified into two groups: time domain and frequency domain measures. The time-domain rhythm measures include temporal metrics like vocalic and non-vocalic durations and their derivatives, etc, and the frequency-domain rhythm measures comprise amplitude-modulated rhythm formant trajectory frequency and magnitude. To conduct this research, we focused on four Assamese regional varieties spoken in four districts of the Indian state of Assam. Data used in this study consists of read speech data of native Assamese speakers reading the Assamese translation of the "North Wind and the Sun" passage. The speech data was obtained from a total of 10 speakers for each of the four varieties. The average accuracy of dialect classification using quadratic discriminant analysis turns out to be 42% and 35%, respectively, for time- and frequency-domain rhythm measures.