导入numpy模块后调用input()报TypeError问题求助

• 报错和numpy导入逻辑无关，核心是你把Python内置的input函数名当成自定义变量名用了：代码中input = np.sin(2*np.pi*np.arange(fs*duration)*f/fs)这行执行后，input标识符就不再是用来获取用户输入的内置函数，而是指向你生成的numpy音频数组。
• 如果你在Jupyter、IPython这类会保留运行状态的交互环境调试，只要之前运行过上面那行赋值语句，哪怕你把n = input("Give")写在赋值行前面，重跑代码时内核里的input已经被替换成ndarray对象，调用时必然抛出TypeError: 'numpy.ndarray' object is not callable错误。
• 你想的3种隔离方案都能绕开问题，但完全没必要搞这么复杂，最直接的修复方式就是不要占用内置函数名。

把占用内置名的变量、函数参数全部改名即可，改完直接运行：

import pyaudio
import numpy as np

noteslist = {"C":-10,"C#":-9,"D":-8,"D#":-7,"E":-6,"E#":-5,"F":-4,"F#":-3,"G":-2,"G#":-1,"A":0,"A#":1,"B":2}

n = input("Give str") 

fs = 44100       
duration = 5.0  
f = 432.0        

# 原变量名input改为audio_signal，避免覆盖内置函数
audio_signal = np.sin(2*np.pi*np.arange(fs*duration)*f/fs)

# 参数名也不要用内置函数名
def play(signal):
    sample = signal.astype(np.float32)
    volume = 0.25
    p = pyaudio.PyAudio()
    stream = p.open(format=pyaudio.paFloat32, channels=1, rate=fs, output=True)
    stream.write(volume*sample)
    stream.stop_stream()
    stream.close()
    p.terminate()

play(audio_signal)

交互环境调试的话，改完代码一定要重启内核再运行，重置之前被覆盖的内置input函数，否则报错不会消失。

如果确实需要做逻辑隔离，可按下面方式实现，不需要改内置名也能跑：

• 方案1：拆分独立文件
  新建audio_utils.py存放音频相关逻辑，和输入逻辑完全隔离：

  # audio_utils.py 内容
  import pyaudio
  import numpy as np
  
  def play_sine_wave(freq=432.0, sample_rate=44100, duration=5.0, volume=0.25):
      wave = np.sin(2*np.pi*np.arange(sample_rate*duration)*freq/sample_rate)
      sample = wave.astype(np.float32)
      p = pyaudio.PyAudio()
      stream = p.open(format=pyaudio.paFloat32, channels=1, rate=sample_rate, output=True)
      stream.write(volume*sample)
      stream.stop_stream()
      stream.close()
      p.terminate()
  

  主文件只处理输入，需要播放时再导入模块：

  # 主文件内容
  noteslist = {"C":-10,"C#":-9,"D":-8,"D#":-7,"E":-6,"E#":-5,"F":-4,"F#":-3,"G":-2,"G#":-1,"A":0,"A#":1,"B":2}
  n = input("Give str")
  # 输入全部处理完再导入音频模块
  from audio_utils import play_sine_wave
  play_sine_wave()
  

• 方案2、3：分段隔离/延后导入numpy
  把所有需要调用input()获取用户输入的逻辑放在最前面，全部跑完再导入numpy、写音频生成逻辑，避免提前覆盖标识符：

  # 第一部分：不导入numpy，先处理所有输入
  noteslist = {"C":-10,"C#":-9,"D":-8,"D#":-7,"E":-6,"E#":-5,"F":-4,"F#":-3,"G":-2,"G#":-1,"A":0,"A#":1,"B":2}
  n = input("Give str")
  fs = 44100       
  duration = 5.0  
  f = 432.0        
  
  # 第二部分：输入全部获取完成，再导入依赖运行音频逻辑
  import pyaudio
  import numpy as np
  audio_signal = np.sin(2*np.pi*np.arange(fs*duration)*f/fs)
  
  def play(signal):
      sample = signal.astype(np.float32)
      volume = 0.25
      p = pyaudio.PyAudio()
      stream = p.open(format=pyaudio.paFloat32, channels=1, rate=fs, output=True)
      stream.write(volume*sample)
      stream.stop_stream()
      stream.close()
      p.terminate()
  
  play(audio_signal)
  

写Python代码时尽量不要用input、list、dict、str这类内置函数/类名当自定义变量名，这类问题排查起来非常浪费时间。

内容的提问来源于stack exchange，提问作者Martin