Closed dodobyte closed 4 years ago
I plot the embedding vector and it's mostly zeros. Is this expected?
I want to use the embedding in another project. I also plotted their example embeddings and those seem to be distributed significantly better.
Embedding from this project
Embeddings from AutoVC
And here's the test code;
from pathlib import Path from resemblyzer import VoiceEncoder, preprocess_wav import numpy as np, matplotlib.pyplot as plt wav = preprocess_wav(Path("367-130732-0005.flac")) encoder = VoiceEncoder() embed = encoder.embed_utterance(wav) plt.plot(embed, 'bo') plt.show()
Am I doing something wrong? Thanks.
Absolutely, the embeddings are sparse due to the relu at the end of the model. It doesn't make them worse, although I did remove that relu in the development branch I'm working on. Don't worry about it.
I plot the embedding vector and it's mostly zeros. Is this expected?
I want to use the embedding in another project. I also plotted their example embeddings and those seem to be distributed significantly better.
Embedding from this project
Embeddings from AutoVC
And here's the test code;
Am I doing something wrong? Thanks.