VoiceSplit icon indicating copy to clipboard operation
VoiceSplit copied to clipboard

Example 5 import version mismatch?

Open andresrivero opened this issue 2 years ago • 1 comments

I was trying to load example # 5 in the readme locally, everything goes fine until the section after importing import sys sys.path.insert(0, "./VoiceSplit/") sys.path.insert(0, "./GE2E-Speaker-Encoder/")

Maybe GE2E was updated after this was originally used, I couldn't find where in the version it fit. Just loading from the master branch causes an error like this:

Testing the encoder... Traceback (most recent call last): File "C:\Users\Studio\dev\voice\v.py", line 42, in <module> embed = encoder.embed_utterance(wav) File "C:\Users\Studio\dev\voice\./GE2E-Speaker-Encoder\encoder\inference.py", line 145, in embed_utterance frames = audio.wav_to_mel_spectrogram(wav) File "C:\Users\Studio\dev\voice\./GE2E-Speaker-Encoder\encoder\audio.py", line 48, in wav_to_mel_spectrogram frames = librosa.feature.melspectrogram( TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given

I'm not familiar with the GE2E branch to revert or modify it to be able to run. Any help would be appreciated :)

andresrivero avatar Mar 28 '23 17:03 andresrivero

I was trying to load example # 5 in the readme locally, everything goes fine until the section after importing import sys sys.path.insert(0, "./VoiceSplit/") sys.path.insert(0, "./GE2E-Speaker-Encoder/")

Maybe GE2E was updated after this was originally used, I couldn't find where in the version it fit. Just loading from the master branch causes an error like this:

Testing the encoder... Traceback (most recent call last): File "C:\Users\Studio\dev\voice\v.py", line 42, in <module> embed = encoder.embed_utterance(wav) File "C:\Users\Studio\dev\voice\./GE2E-Speaker-Encoder\encoder\inference.py", line 145, in embed_utterance frames = audio.wav_to_mel_spectrogram(wav) File "C:\Users\Studio\dev\voice\./GE2E-Speaker-Encoder\encoder\audio.py", line 48, in wav_to_mel_spectrogram frames = librosa.feature.melspectrogram( TypeError: melspectrogram() takes 0 positional arguments but 2 positional arguments (and 2 keyword-only arguments) were given

I'm not familiar with the GE2E branch to revert or modify it to be able to run. Any help would be appreciated :)

Downgrade librosa to 0.9.1 or 0.9.2

pip install librosa==0.9.1

justinjohn0306 avatar Apr 06 '23 01:04 justinjohn0306