在艺术嗓音客观评价中,评价声学参数的选取至关重要。尝试利用语音分析技术提取歌声平均能量、平均频率误差、平均音域误差作为客观评价声学参数,使用神经网络方法客观评价艺术嗓音歌声质量,并与资深专业教师的主观评价进行比较。结果表明,综合使用三个声学参数能正确客观评价歌声质量,评价误差在4%之内,有助于科学地指导选拔、培养和训练艺术嗓音人才。 In the objective evaluation of artistic voice, the evaluation of acoustic parameters is of crucial importance. Attempt to extract average voice energy, average frequency error and average range error as the objective evaluation acoustic parameters by using speech analysis technology, and use the neural network method to objectively evaluate the voice quality of art voice, and compare it with the subjective evaluation of senior professional teachers. The results show that the comprehensive use of three acoustic parameters can correctly and objectively evaluate the singing voice quality, and the evaluation error is within 4%, which helps to guide, select, train and train art voice professionals.