一种高精度改进型SHR基音检测算法

利用正弦语音模型中浊音存在的谐波与子谐波,在SHR(subharninctoharmonicratio)算法的基础上,提出了一种改进型高精度基音检测算法ISHR(improvingsubharninctoharmonicratio)。根据幅度调制和频率调制在语音分析中的特性、频域中幅度值和自相关频率比值,该方法采用基于正弦模型的均方误差对语音进行检测,提取出准确基音。仿真结果表明此种算法在基音提取中具有高精度及高可靠性。

Improved SHR pitch detection algorithm with high accuracy

An improving sub-harmonic to harmonic ratio(ISHR)algorithm based on sub-harmonic to harmonic ratio(SHR)and using the voiced speech harmonic and sub-harmonic in the speech sinusoidal model algorithm was proposed.According to the characteristics of amplitude and frequency modulation in speech analysis,amplitude and autocorrelation ratio in frequency domain,this algorithm through the mean square error obtained from the sinusoidal model could detect accurate pitch.Simulation results indicate that the proposed algorithm possesses high accuracy and reliability in speech pitch detection.