Speech Name Database Establishing Of Intelligence Telephone-based Military Recruiting System

In recent years, speech synthesis is a heated research aspect in the field of speech signal processing and a hotspot in the artificial intelligence. It is popularly used in many fields, which improves human-computer interaction and brings great social benefits to us. In the first part of this paper is the introduction of research background of speech synthesis technique and historical review of speech synthesis and text to speech systems. Afterwards we give an analysis of several classical speech synthesis methods.In all fields of the digital processing of speech signal, for instance, speech analysis, speech synthesis, speech compression coding and speech recognition with confirm by speaker, etc, detecting pitch period accurately and reliably is extremely important, which will influence the performance of the whole system directly. It is difficult to synthesize the sonant in the speech synthesis especially in the synthesizing diphthong in speech synthesis. A kind of effective method for getting pitch period of voice id initiated in this paper. We adopt the exciting VTFR method to accomplish the speech synthesis of voice. Outline is as follows;(1) A kind of effective method is initiated in this paper and it is to get pitch period of voice by the means of displacement matching in segment. This can prevent effectively judging two periods of cycle by mistake because of two similar wave in one period of cycle. Prevent the possibility of getting the times periods of cycle.(2) Has done the detailed research to voice synthesis in the sonant and accomplished the whole word synthesis. Deeply analyse about characteristics of Chinese surd and sonant and explain the synthesis of surd and sonant in detail respectively. Has introduced the connect of surd and sonant detailedly(3) With the algorithm in this paper, we have accomplished speech name database of intelligence telephone-based military recruiting system. Simulation experiments showed that the proposed algorithms have high accuracy in pitch estimate, strong robustness and high applied value.