This work is concerned with the feedback stabilization of the following semilinear system with distributed delay \begin{equation*} \begin{array}{c}\dot{w}(t)=\mathcal {A} w(t)+v(t)F(w_{t})\ \text{and}\ w_{0}=\phi \in \mathcal {C}:=C([-r,0]; \mathcal {X}), \end{array} \tag{1} \end{equation*} where $\mathcal {A}$ is the infinitesimal generator of a $C_{0}-$ semigroup $(S(t))_{t\geq 0}$ on a Hilbet space $\mathcal {X}$ . $F$ is a locally Lipschitz function from $\mathcal {C}$ to $\mathcal {X}$ such that $F(0)=0$ and $\mathcal {C}$ is the space of all continuous function from $[-r,0]$ to $\mathcal {X}$ which is equipped with the following norm $\Vert \varphi \Vert _{\mathcal {C}}=\displaystyle \sup _{-r\leq \theta \leq 0} \Vert \varphi (\theta)\Vert _{\mathcal {X}}$ for all $\varphi \in \mathcal {C}$ . The contro l $v$ , of type feedback, is a defined function from $\mathbb {R}^{+}$ to $\mathbb {R}.$ For $t\geq 0$ , the history function $w_{t} \in \mathcal {C}$ is defined by \begin{equation*} w_{t}(\theta)=w(t+\theta)\ \; \text{for all}\; \theta \in [-r,0]. \end{equation*} Here, we consider an observation condition in terms of the semigroup solution of system ( ) with $F=0$ and parameterized by a positive constant $\gamma$ , it means that there exist $T,\,\delta >0$ and $\gamma \geq 2$ such that \begin{equation*} \int _{0}^{T}|\langle F(U(s)\xi), (U(s)\xi)(0)\rangle | ds \geq \delta \Vert \xi (0)\Vert _{\mathcal {X}}^{\gamma } \;\;\;\; \text{for all}\; \xi \; \in \;\mathcal {C}. \end{equation*} Some sufficient conditions are given with respect to a bounded feedback control and the parameter $\gamma$ to guarantee the feedback stabilization of the semilinear system with distributed delay. Moreover, for $\gamma \geq 2,$ an explicit polynomial decay rate of the stabilization state is estimated in the strong stabilization case, more precisely, we show that \begin{equation*} \left\Vert w(t)\right\Vert _{\mathcal {X}}=O\left(t ^{-\dfrac{1}{2(\gamma -1)}}\right) \text{ as } t \longrightarrow +\infty. \, \tag{2} \end{equation*} In the bilinear case with distributed delay and by using the decomposition method of the state space $\mathcal {X}$ , we investigate the feedback stabilization of system (1) using some suitable conditions like observability assumption. In the case of strong stabilization, we obtain the same explicit decay estimate (2) of the stabilized state. Furthermore, when $\gamma = 2$ , we show that the normalized feedback control exponentially stabilizes the semilinear system (1). The obtained results are illustrated by three examples and numerical simulations for wave equation with distributed delay.