Python ValueError:操作数无法与形状（1024，）（1024,2）一起广播

唉，我知道这与其他几个问题类似，但我这里的问题是我不太理解specgram函数在Matlab中是如何工作的。我的代码基于以下示例：

作为回报，我收到

ValueError:操作数不能与形状（1024，）（1024,2）

一起广播。我在这里不明白什么

编辑：完全回溯：

ValueError                                Traceback (most recent call last)
<ipython-input-13-4952b82b74f2> in <module>()
     16 subplot(212, sharex=ax1)
     17 Pxx, freqs, bins, im = specgram(x, NFFT=NFFT, Fs=Fs, noverlap=900,
---> 18                                 cmap=cm.gist_heat)
     19 show()
     20 

/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/pyplot.pyc in specgram(x, NFFT, Fs, Fc, detrend, window, noverlap, cmap, xextent, pad_to, sides, scale_by_freq, hold, **kwargs)
   3145                           window=window, noverlap=noverlap, cmap=cmap,
   3146                           xextent=xextent, pad_to=pad_to, sides=sides,
-> 3147                           scale_by_freq=scale_by_freq, **kwargs)
   3148         draw_if_interactive()
   3149     finally:

/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/axes.pyc in specgram(self, x, NFFT, Fs, Fc, detrend, window, noverlap, cmap, xextent, pad_to, sides, scale_by_freq, **kwargs)
   8932 
   8933         Pxx, freqs, bins = mlab.specgram(x, NFFT, Fs, detrend,
-> 8934              window, noverlap, pad_to, sides, scale_by_freq)
   8935 
   8936         Z = 10. * np.log10(Pxx)

/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/mlab.pyc in specgram(x, NFFT, Fs, detrend, window, noverlap, pad_to, sides, scale_by_freq)
    467 
    468     Pxx, freqs, t = _spectral_helper(x, x, NFFT, Fs, detrend, window,
--> 469         noverlap, pad_to, sides, scale_by_freq)
    470     Pxx = Pxx.real #Needed since helper implements generically
    471 

/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/mlab.pyc in _spectral_helper(x, y, NFFT, Fs, detrend, window, noverlap, pad_to, sides, scale_by_freq)
    262     for i in range(n):
    263         thisX = x[ind[i]:ind[i]+NFFT]
--> 264         thisX = windowVals * detrend(thisX)
    265         fx = np.fft.fft(thisX, n=pad_to)
    266 

ValueError: operands could not be broadcast together with shapes (1024,) (1024,2) 

ValueError回溯（最近一次调用）
在（）
16子批次（212，sharex=ax1）
17 Pxx，频率，箱，im=specgram（x，NFFT=NFFT，Fs=Fs，noverlap=900，
--->18 cmap=厘米（热量）
19 show（）
20
/specgram中的Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/pyplot.pyc
3145 window=window，noverlap=noverlap，cmap=cmap，
3146 xextent=xextent，pad_to=pad_to，sides=sides，
->3147按频率缩放=按频率缩放，**kwargs）
3148 draw_if_interactive（）
3149最后：
/specgram中的Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/axes.pyc（self、x、NFFT、Fs、Fc、detrend、window、noverlap、cmap、xextent、pad_to、sides、scale_by_freq、**kwargs）
8932
8933 Pxx，频率，箱=mlab.specgram（x，NFFT，Fs，detrend，
->8934窗户，noverlap，墙板，侧面，按频率缩放）
8935
8936 Z=10。*np.log10（Pxx）
/specgram中的Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/mlab.pyc（x、NFFT、Fs、detrend、window、novelap、pad_to、sides、scale_by_freq）
467
468 Pxx，freqs，t=_spectrum_helper（x，x，NFFT，Fs，detrend，window，
-->469十一月地图，地图，侧面，按频率缩放）
470 Pxx=Pxx.real#需要，因为helper一般实现
471
/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/mlab.pyc in_spectrum_helper（x、y、NFFT、Fs、detrend、window、novelap、pad_to、sides、scale_by_freq）
262对于范围内的i（n）：
263 thisX=x[ind[i]：ind[i]+NFFT]
-->264 thisX=窗口VAL*detrend（thisX）
265 fx=np.fft.fft（thisX，n=pad_to）
266
ValueError:操作数无法与形状（1024，）（1024,2）一起广播

这不是代码的问题：而是WAV文件的问题。该文件有两个音频通道，因此位于数据的第二行。

您能发布完整的异常回溯吗？我假设这里的错误在specgram中得到提升？添加了完整的回溯。啊！有趣。这似乎是我的WAV文件的问题？奇怪，因为我刚刚让iTunes为我转换了…嗯。。我认为问题在于立体声文件。我有第二个，现在，由大胆创造。。。

#!/usr/bin/env python

from pylab import *
import scipy.io.wavfile

dt = 0.0005

sr, x = scipy.io.wavfile.read(fname) # the signal
NFFT = 1024       # the length of the windowing segments
Fs = int(1.0/dt)  # the sampling frequency

# Pxx is the segments x freqs array of instantaneous power, freqs is
# the frequency vector, bins are the centers of the time bins in which
# the power is computed, and im is the matplotlib.image.AxesImage
# instance

ax1 = subplot(211)
plot(x)
subplot(212, sharex=ax1)
Pxx, freqs, bins, im = specgram(x, NFFT=NFFT, Fs=Fs, noverlap=900,
                                cmap=cm.gist_heat)
show()

ValueError                                Traceback (most recent call last)
<ipython-input-13-4952b82b74f2> in <module>()
     16 subplot(212, sharex=ax1)
     17 Pxx, freqs, bins, im = specgram(x, NFFT=NFFT, Fs=Fs, noverlap=900,
---> 18                                 cmap=cm.gist_heat)
     19 show()
     20 

/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/pyplot.pyc in specgram(x, NFFT, Fs, Fc, detrend, window, noverlap, cmap, xextent, pad_to, sides, scale_by_freq, hold, **kwargs)
   3145                           window=window, noverlap=noverlap, cmap=cmap,
   3146                           xextent=xextent, pad_to=pad_to, sides=sides,
-> 3147                           scale_by_freq=scale_by_freq, **kwargs)
   3148         draw_if_interactive()
   3149     finally:

/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/axes.pyc in specgram(self, x, NFFT, Fs, Fc, detrend, window, noverlap, cmap, xextent, pad_to, sides, scale_by_freq, **kwargs)
   8932 
   8933         Pxx, freqs, bins = mlab.specgram(x, NFFT, Fs, detrend,
-> 8934              window, noverlap, pad_to, sides, scale_by_freq)
   8935 
   8936         Z = 10. * np.log10(Pxx)

/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/mlab.pyc in specgram(x, NFFT, Fs, detrend, window, noverlap, pad_to, sides, scale_by_freq)
    467 
    468     Pxx, freqs, t = _spectral_helper(x, x, NFFT, Fs, detrend, window,
--> 469         noverlap, pad_to, sides, scale_by_freq)
    470     Pxx = Pxx.real #Needed since helper implements generically
    471 

/Library/Python/2.7/site-packages/matplotlib-1.3.1-py2.7-macosx-10.8-x86_64.egg/matplotlib/mlab.pyc in _spectral_helper(x, y, NFFT, Fs, detrend, window, noverlap, pad_to, sides, scale_by_freq)
    262     for i in range(n):
    263         thisX = x[ind[i]:ind[i]+NFFT]
--> 264         thisX = windowVals * detrend(thisX)
    265         fx = np.fft.fft(thisX, n=pad_to)
    266 

ValueError: operands could not be broadcast together with shapes (1024,) (1024,2)