Design And Implementation Of Automatic Tuning System Based On SoC

The daily use of stringed instruments is inseparable from the tuner.A tuner is a tool that helps people correct the pitch of an instrument.At present,most of the tuner on the market can only help users to obtain the current pitch of the string.The function of the tuner is so single that can not meet the user's tuning needs.The specificity of the tuner is also too strong.Therefore,designing a tuning system that can not only be suitable for a variety of stringed instruments but also can quickly and accurately automatic tuning has practical value.The thesis uses DE10-Nano-SoC development board as the core processing platform of the system.The thesis implements a multi-functional tuning system that integrates automatic tuning,automatic winding,wireless control and metronome by combining core processing platform,various hardware function modules and a smart phone.The system hardware design mainly includes the following aspects.In order to realize the acquisition of the sound signal,the thesis designs the audio acquisition module.And the thesis implements the tuning control module to realize the automatic tuning function.For enhancing the human-computer interaction experience,the thesis designs the voice broadcast module.The thesis accomplishes the network communication module to realize the mobile terminal communication with the embedded platform.Finally,the thesis uses SoC technology to integrate the above hardware function modules to complete the overall hardware design of the system.The main work of system software design are as follows.The thesis uses the improved endpoint detection algorithm to achieve the separation of string and non-string sounds.And the thesis uses the improved pitch detection algorithm to realize the accurate detection of the pitch of the string.The thesis uses the motor control method which combines coarse adjustment with fine adjustment to achieve fast tuning.Finally,the thesis develops the mobile phone software to facilitate the user's control of the tuning system and the daily rhythm practice.The main contributions of the thesis are as follows:(1)For the problem that the traditional endpoint detection method is vulnerable to environmental impact,the thesis uses the dynamic threshold update method to improve the endpoint detection algorithm according to the characteristics of the string signal,which improves the accuracy of the endpoint detection algorithm to some extent.(2)Aiming at the problem that a single algorithm is difficult to accurately detect the pitch frequency in a wide frequency range,the thesis adopts the method of segmentation detection and uses different algorithms to detect the pitch in different frequency ranges.(3)For the problem that the autocorrelation algorithm and the harmonic peak method are easy to produce the misjudgment of double frequency,the thesis uses the target frequency to estimate the pitch frequency range to avoid the misjudgment of double frequency and also reduce the calculation amount of the algorithm to some extent.(4)In order to improve the accuracy of the harmonic peak detection,the thesis uses the three-line ratio method to correct the fundamental frequency.After several tests,the automatic tuning system designed by the thesis is stable in operation and has a wide tuning range.It can achieve tuning accuracy within ±5 cents in the frequency range of 40 Hz to 800 Hz,and satisfy the tuning needs of various string instruments such as guitar and ukulele.The tuning time of the single string is less than 10 s and all performance indicators of the system have reached the expected research target.