Error when using Speaker Diarization pipeline
thangld201 opened this issue · 2 comments
thangld201 commented
When I run the following command (as illustrated in the tutorial !python audioAnalysis.py speakerDiarization -i ./data/diarizationExample.wav --num 2 --flsd
. I got the error
Traceback (most recent call last):
File "audioAnalysis.py", line 617, in <module>
speakerDiarizationWrapper(args.input, args.num, args.flsd)
File "audioAnalysis.py", line 237, in speakerDiarizationWrapper
aS.speaker_diarization(inputFile, numSpeakers, lda_dim=5, plot_res=True)
File "/content/pyAudioAnalysis/pyAudioAnalysis/audioSegmentation.py", line 832, in speaker_diarization
at.load_model(os.path.join(base_dir, "svm_rbf_speaker_10"))
File "/content/pyAudioAnalysis/pyAudioAnalysis/../pyAudioAnalysis/audioTrainTest.py", line 530, in load_model
mean = cPickle.load(fo)
ValueError: unsupported pickle protocol: 5
I'm using google colaboratory environment and already installed the libraries and dependencies, what could be the cause ?
thangld201 commented
@tyiannak Could you help me please ?
thangld201 commented
I found where the error came from: Google Colab uses pickle 4 by default, so you have to do
!pip install pickle5
import pickle5 as cPickle