顺利地“Audialize”在Python中与Turtle和PyAudio一起递归

Question

我希望能够重现下面代码背后的想法，但是听起来并不可怕。理想情况下，我想要一个使用学校孩子可以理解的代码的解决方案（可以导入play_note函数，这样他们就不必担心它是如何工作的）。一个回答here建议连续音符之间的点击是由于声音的不完整周期，但我不知道如何解决改变持续时间。

任何人都可以帮忙吗？它可以使一些调整工作，或者是某种方式有缺陷吗？

import turtle 
import pyaudio 
import numpy as np 

def play_note(freq, dur): 

    p = pyaudio.PyAudio() 

    volume = 0.5  # range [0.0, 1.0] 
    fs = 44100  # sampling rate, Hz, must be integer 
    duration = dur # in seconds, may be float 
    f = freq  # sine frequency, Hz, may be float 

    # generate samples, note conversion to float32 array 
    samples = (np.sin(2*np.pi*np.arange(fs*duration)*f/fs)).astype(np.float32) 

    # for paFloat32 sample values must be in range [-1.0, 1.0] 
    stream = p.open(format=pyaudio.paFloat32, 
        channels=1, 
        rate=fs, 
        output=True) 

    # play. May repeat with different volume values (if done interactively) 
    stream.write(volume*samples) 

    stream.stop_stream() 
    stream.close() 

    p.terminate() 

def tree(branchLen,t): 
    if branchLen > 5: 
     freq = branchLen * 2 + 400 
     dur = branchLen/100.0 
     print freq, dur 
     play_note(freq, dur) 
     t.forward(branchLen) 
     t.right(20) 
     tree(branchLen-15,t) 
     t.left(40) 
     tree(branchLen-15,t) 
     t.right(20) 
     t.backward(branchLen) 

def main(): 
    t = turtle.Turtle() 
    t.speed(0) 
    myWin = turtle.Screen() 
    t.left(90) 
    t.up() 
    t.backward(100) 
    t.down() 
    t.color("green") 
    tree(75,t) 
    myWin.exitonclick() 

main() 

Answer 1

我不确定你到底需要什么。我试图运行你的代码。它遇到了一些问题。所以我修好了。然后汇集所有变量以便于访问。修改变量的函数，以便可以从主函数中设置它们。导入所需的模块。

我在递归之外提供了freq和duration。因此它们是固定的。在原始代码中，freq增加并且持续时间相对于树长度减少。这是噪音的原因之一，因为freq和duration然后不是60秒的倍数，并因此产生导致噪音的差距。使用固定的freq和duration，噪音最小。

由于创建新分支和递归启动需要几微秒的延迟，会有一些噪音。 这里是EDITED代码。

import pyaudio 
import turtle 
import numpy as np 

def play_note(freq, dur): 

    p = pyaudio.PyAudio() 

    volume = 0.5  # range [0.0, 1.0] 
    fs = 44100  # sampling rate, Hz, must be integer 
    duration = dur # in seconds, may be float 
    f = freq  # sine frequency, Hz, may be float 

    # generate samples, note conversion to float32 array 
    samples = (np.sin(2*np.pi*np.arange(fs*duration)*f/fs)).astype(np.float32) 

    # for paFloat32 sample values must be in range [-1.0, 1.0] 
    stream = p.open(format=pyaudio.paFloat32, 
        channels=1, 
        rate=fs, 
        output=True) 

    # play. May repeat with different volume values (if done interactively) 
    stream.write(volume*samples) 

    stream.stop_stream() 
    stream.close() 

    p.terminate() 

def tree(branchLen,t): 
    if branchLen > 5: 

     freq = branchLen * 2 + 420 
     dur = branchLen/60.0 
     print "branchLen={} , Freq = {}, Duration={}".format(branchLen, freq, dur) 
     t.forward(branchLen) 
     play_note(freq, dur) 
     t.right(20) 
     tree(branchLen-15,t) 
     play_note(freq, dur) 
     t.left(40) 
     play_note(freq, dur) 
     tree(branchLen-15,t) 
     play_note(freq, dur) 
     t.right(20) 
     play_note(freq, dur) 
     t.backward(branchLen) 
     play_note(freq, dur) 


def main(): 
    t = turtle.Turtle() 
    t.speed(turtle_speed) 
    myWin = turtle.Screen() 
    t.left(branch_left_turn) 
    t.up() 
    t.backward(branch_back_turn) 
    t.down() 
    t.color(tree_color) 
    tree(tree_length,t) 
    myWin.exitonclick() 

if __name__ == '__main__': 
    #Global Variables brought together for easy setting 
    turtle_speed = 0.5 #speed of turtle graphics <1 -> faster ; >1 -> slower 
    tree_length = 60 #keep within multiples of 60 for smoothness 
    branch_left_turn = 90 
    branch_back_turn = 100 
    tree_color = "green" #can experiment with "red" "blue" etc 
    #Call main function 
    main() 

输出数据：

Python 2.7.9 (default, Dec 10 2014, 12:24:55) [MSC v.1500 32 bit (Intel)] on win32 
Type "copyright", "credits" or "license()" for more information. 
>>> ================================ RESTART ================================ 
>>> 
branchLen=60 , Freq = 540, Duration=1.0 
branchLen=45 , Freq = 510, Duration=0.75 
branchLen=30 , Freq = 480, Duration=0.5 
branchLen=15 , Freq = 450, Duration=0.25 
branchLen=15 , Freq = 450, Duration=0.25 
branchLen=30 , Freq = 480, Duration=0.5 
branchLen=15 , Freq = 450, Duration=0.25 
branchLen=15 , Freq = 450, Duration=0.25 
branchLen=45 , Freq = 510, Duration=0.75 
branchLen=30 , Freq = 480, Duration=0.5 
branchLen=15 , Freq = 450, Duration=0.25 
branchLen=15 , Freq = 450, Duration=0.25 
branchLen=30 , Freq = 480, Duration=0.5 
branchLen=15 , Freq = 450, Duration=0.25 
branchLen=15 , Freq = 450, Duration=0.25 
>>> 

输出图像时的振幅急剧变化，例如，当一个正弦波的一个的峰值附近打断

Answer 2

次数发生在音频其周期。解决方案是平滑淡出（可能是正弦波）。通常，当合成声音时，这些音量变化称为声音的包络线。

为您的代码的简单（但可能不是很有效）信封看起来是这样的：

def play_note(freq, dur): 

    p = pyaudio.PyAudio() 

    volume = 0.5  # range [0.0, 1.0] 
    fs = 44100  # sampling rate, Hz, must be integer 
    duration = dur # in seconds, may be float 
    f = freq  # sine frequency, Hz, may be float 

    # generate samples, note conversion to float32 array 
    samples = (np.sin(2*np.pi*np.arange(fs*duration)*f/fs)).astype(np.float32) 

    #Fade out the end 
    release_time = 0.020 #(seconds) 
    release_samples = np.ceil(release_time * fs) 
    fade_curve = np.linspace(1.0, 0.0, num=release_samples) 
    samples[-release_samples:] *= fade_curve 

这可以通过改变释放时间或使用不同的褪色曲线进行调整。它也可以扩展为在整个笔记上应用ADSR信封。