It is well known that determining the structure and training the parameters of neural networks efficiently are difficult in the field of neural networks research. These years it has been somewhat successful to construct neural networks in light of Bayesian theorem and to optimize neural networks according to particle swarm optimization respectively. A novel model of taste signals recognition based on minimal uncertainty neural networks is proposed in this paper. The model adopts minimization uncertainty adjudgment to construct the networks structure, and uses Bayesian theorem and particle swarm optimization (PSO) to determine the parameters of the networks rapidly and efficiently. The identification of the taste signals of 10 kinds of tea is successful in utilization of this model. The experimental results show the feasibility and probability of introducing the proposed model to the identification of taste signals of tea. Section 2 presents the model of minimal uncertainty neural networks (MUNN). How to determine the weights and biases of MUNN by Bayesian theorem, PSO, and the hybrid of them are illustrated respectively in section 3. The experimental results are presented and discussed in section 4. Conclusions are in section 5.