This study formulated a surface EMG model capable of representing the variance distribution of EMG signals. In the model, EMG signals are handled based on a Gaussian white noise process with a mean of zero for each variance value. EMG signal variance is taken as a random variable that follows an inverse gamma distribution, allowing the representation of noise superimposed onto this variance. The estimation of the variance distribution based on marginal likelihood maximization was also proposed. The procedure can be approximated using rectified and smoothed EMG signals, thereby allowing the determination of distribution parameters in real time at low computational cost. The variance distribution estimated using the proposed model exhibits potential for the estimation of muscle force.