Pablo Alonso palonso
palonso
/ extract-openl3-embeddings.py
Last active
August 31, 2021 09:14
Extract Openl3 embeddings in Essentia.
View extract-openl3-embeddings.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from pathlib import Path | |
| import essentia.standard as es | |
| import numpy as np | |
| from essentia import Pool | |
| class MelSpectrogramOpenL3: | |
| def __init__(self, hop_time): | |
| self.hop_time = hop_time |
palonso
/ ismir_lbd_deep-embeddings-with-essentia-models.ipynb
Last active
September 29, 2020 10:04
Embedding extraction demo for the ISMIR 2020 LBD "Deep embeddings with Essentia models"
View ismir_lbd_deep-embeddings-with-essentia-models.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
palonso
/ freezer.py
Last active
February 11, 2022 08:15
Serialize a TensorFlow graph as a Protobuf file. Remove the nodes not into the `model` namespace.
View freezer.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # tensorflow: loading model | |
| sess = tf.Session() | |
| sess.run(tf.global_variables_initializer()) | |
| saver = tf.train.Saver() | |
| results_folder = model + '/' | |
| saver.restore(sess, results_folder) | |
| gd = sess.graph.as_graph_def() |
palonso
/ nsgcq_encoded.py
Last active
November 14, 2019 17:12
Forward and Backward NSGCQ transforms. Complex values are encoded in 4 channes: real part magnitude, imag part magnitude, real part sign and imag part sign. This representation seems suitable for feeding NNs while allowing perfect reconstruction.
View nsgcq_encoded.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import numpy as np | |
| import matplotlib.pyplot as plt | |
| import essentia.pytools.spectral as sp | |
| from essentia.standard import MonoLoader, MonoWriter | |
| from essentia import lin2db | |
| PARAMS = { | |
| 'frameSize': 2 ** 14, | |
| 'sampleRate': 44100, |
palonso
/ real-time_auto-tagging_with_essentia_tensorflow_pyaudio_musicnn.ipynb
Last active
October 28, 2019 16:09
Real-time auto-tagging example using Essentia with Tensorflow and PyAudio to process the soundcard input using the MusiCNN model.
View real-time_auto-tagging_with_essentia_tensorflow_pyaudio_musicnn.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.