We present an overview of different strategies and refinements to share parameters in HMM models at distribution (state) level for continuous speech recognition, showing the advantages and drawbacks of the different kinds of modeling [6]. We compare them with sharing at model level [5], achieving an error reduction close to 20% [4]. Discrete, semi continuous and continuous HMM models are also compared using these approaches. We consider two ways to smooth discrete distributions (interpolate detailed context dependent with robust context independent) derived from deleted interpolation [1] and coocurrence smoothing [10].