audeering / opensmile-python

Python package for openSMILE
https://audeering.github.io/opensmile-python/
Other
243 stars 32 forks source link

The output results of process_file and process_signal are inconsistent #44

Closed srdfjy closed 3 years ago

srdfjy commented 3 years ago

Hi,I found that with the same configuration and the same wav file, the output results of process_file and process_signal are inconsistent.

code

import scipy.io.wavfile as wf

y = smile.process_file('test.wav') sr, signal = wf.read("test.wav") y1 = smile.process_signal(signal, sr) print(y.shape, y.values) print(y1.shape, y1.values)

result

(1, 1582) [[0.23114355 0.69829684 0.373669 ... 0.02352941 5. 4.13 ]] (1, 1582) [[0. 0.21167883 5.900316 ... 0. 7. 4.13 ]]

srdfjy commented 3 years ago

Sorry, it's my problem, it needs to be divided by 32768.