Application of PNN-HMM model based on emotion-speech combination in broadcast intelligent communication analysis

The emotive manifestation of a news anchor ought not to be arbitrary, but rather meticulously crafted and refined elicit more coherent emotional response. Hence, the identification appropriate disposition during broadcasts constitutes meritorious domain inquiry. To address concerns related imprecise extraction speech features intricate detection states in broadcasting, this study presents an innovative Chinese pronunciation system predicated on recognition. GA-SVM algorithm is employed ascertain endpoint input signals extract feature parameters. For recognition temperament broadcast speech, PNN model utilized execute decoding process Viterbi HMM. Experimental findings evince that SVM optimized by GA attains robust classification effect across diverse test samples. Moreover, PNN-HMM exhibits noteworthy capacity withstand noise extraction, thereby enabling accurate discernment characteristics under examination. Additionally, research furnishes valuable point reference for application intelligent technology audio information.