TypeError: mel() takes 0 positional arguments but 2 positional arguments (and 3 keyword-only arguments) were given #25
Description
D:\code\MakeItTalk\main_end2end.py:99: SyntaxWarning: "is not" with a literal. Did you mean "!="?
ains = [item for item in ains if item is not 'tmp.wav']
Loaded the voice encoder model on cuda in 0.01 seconds.
Processing audio file M6_04_16k.wav
0 out of 0 are in this portion
Traceback (most recent call last):
File "D:\code\MakeItTalk\main_end2end.py", line 113, in <module>
au_data_i = c.convert_single_wav_to_autovc_input(audio_filename=os.path.join('examples', ain),
File "D:\code\MakeItTalk\src\autovc\AutoVC_mel_Convertor_retrain_version.py", line 225, in convert_single_wav_to_autovc_input
x_real_src, f0_norm = extract_f0_func_audiofile(audio_file, 'F')
File "D:\code\MakeItTalk\src\autovc\retrain_version\vocoder_spec\extract_f0_func.py", line 97, in extract_f0_func_audiofile
mel_basis = mel(16000, 1024, fmin=90, fmax=7600, n_mels=80).T
TypeError: mel() takes 0 positional arguments but 2 positional arguments (and 3 keyword-only arguments) were given
A error resulted in MakeItTalk\src\autovc\retrain_version\vocoder_spec\extract_f0_func.py module librosa.filters.mel. Solved by changing line 97
mel_basis = mel(16000, 1024, fmin=90, fmax=7600, n_mels=80).T
to
mel_basis = mel(sr=16000, n_fft=1024, fmin=90, fmax=7600, n_mels=80).T
Wish that will take efforts