The traditional English teaching ability assessment algorithm has some problems, such as low accuracy and poor effect. Therefore, this paper proposes an English teaching ability assessment algorithm based on deep learning. According to the proposed method, it analyzes the characteristics of the input layer, convolution layer, pooling layer, and fully connected layer in the deep learning technology. And then, it also help to determine the dimensions of the convolution kernel matrix and derive the weight parameters through back propagation. Finally, it introduces the open-minded computing for deep learning assessment of English teaching ability, calculate the best standard data and best vector data ratio, and establish a deep learning model to realize English teaching ability assessment. The experimental results show that this method can improve the accuracy of English teaching ability assessment.