Question

在这里编写newb尝试首次使用音频（想要制作一个以特定分贝级别切割波形文件的程序）

wf = wave.open('test.wav', 'r')

wfArraydB = []

for i in range(0, wf.getnframes()): # For every frame/sample, this loop is performed.
    waveData = wf.readframes(1) # Can only do 1 frame because struct.unpack can only do 1 frame
    data = struct.unpack("<h", waveData) # Converts wave data from hexadecimal
    a = wf.tell() # Returns the frame that the file pointer is at
    cf = copysign(data[0], 1) # Converts wave data
    #print("data[0]: %d, cf: %d, cf/32767: %d" % (data[0], cf, cf/32767))
    if data[0] != 0: # Eliminates a math module error if the wave sample value is 0
        b = 20 * log10(cf/32767) # Converts non-hexadecimal wave data to dB format
    else:
        b = -90
    #print("frame %d: %d dB" % (a, b))
    wfArraydB.append(int(b))


acceptableLvl = []
acceptableLvlPre = []
threshold = -40

for i in range(0, len(wfArraydB)): 
    if wfArraydB[i] > threshold:
        acceptableLvlPre.append(i)

    if wfArraydB[i] < threshold: 
        if len(acceptableLvlPre) > 4410: # If sequence of frames over -40 is longer than 4410 samples or 0.1s
            acceptableLvl.append(acceptableLvlPre) # Stores frames in which there is sufficiently loud sound
            acceptableLvlPre[:] = [] # Remove all contents from pre-list (using slice assignment)

print(acceptableLvl)

当我进行打印（acceptedLvl）时，我希望看到一个列表列表，它们有两个要求：1。所有值都超过-40（dB，使用音频文件），和2所有列表的长度都超过4410个样本（0.1秒）。

相反，返回的列表包含13个相同列表的重复，只包含12个索引值（这些值都接近~2秒声音文件@ 44,100 hz的末尾）

[[85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039], [85368, 87027, 87028, 87029, 87030, 87031, 87032, 87035, 87036, 87037, 87038, 87039]]

不太明白为什么因为它应该在80,000个样本之前更多地返回值，因为音频文件在开头和中间也有响度（10,000到50,000个样本）

任何有关我的语法或其他任何帮助的帮助，谢谢

Answer 1

Python中的列表是可变的。您的acceptableLvl有13个指向同一列表实例acceptableLvlPre的链接。您修改acceptableLvlPre，因此修改了所有副本。

解决方案很简单：将acceptableLvlPre[:] = []替换为acceptableLvlPre = []。然后会创建一个新列表，一切都会好的。

迭代列表不能正常工作我的想象。为什么呢？

1 个答案: