Xiaoping777
Xiaoping777
I following the clustering / Kmeans method to segment male / female speakers, the performance is pretty bad, not sure if it is the similar issue.
I got the similar issue, then I use: s, fs = librosa.load(audio_file, sr = 16000) to replace the following: fs, s = aIO.read_audio_file(audio_file)
> However, I got another issue after this fix. > > What I found is that after pasting the content into Dash cell, the data_timestamp or data_previous is called twice....