Skip to content

2. Examples for a first try

Jump to bottom
username116 edited this page Aug 27, 2024 · 8 revisions

BPM

  • Create a working folder. Copy an audio file into this folder (flac, mp3, wav...)

  • Using Windows Explorer, in the same folder. Top left menu: File > Open Windows PowerShell

  • Then enter the following commands, one line at a time

    py
    import essentia.standard as es

    Here, replace music.flac with the name and extension of your audio file

    audio = es.MonoLoader(filename="music.flac")()
    rhythm_extractor = es.RhythmExtractor2013(method="multifeature")
    bpm, beats, beats_confidence, _, beats_intervals = rhythm_extractor(audio)
    print("BPM:", bpm)
    Result 

    PS E:\Test> py
Python 3.12.5 (tags/v3.12.5:ff3bc82, Aug  6 2024, 20:45:27) [MSC v.1940 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import essentia.standard as es
2024-08-24 19:12:17.310518: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
[   INFO   ] MusicExtractorSVM: no classifier models were configured by default
2024-08-24 19:12:18.239514: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: SSE SSE2 SSE3 SSE4.1 SSE4.2 AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
>>> audio = es.MonoLoader(filename="09. Slow roll.flac")()
>>> rhythm_extractor = es.RhythmExtractor2013(method="multifeature")
>>> bpm, beats, beats_confidence, _, beats_intervals = rhythm_extractor(audio)
>>> print("BPM:", bpm)
BPM: 156.33270263671875
>>>
>>>
>>> exit()
PS E:\Test>

  • You can exit from Python with exit() or quit() or Ctrl + Z

  • You can clear the screen with Clear-Host or cls


Mood Happy with a model

Still in a working folder with an audio file

  • Download these files:
    mood_happy-discogs-effnet-1.pb
    mood_happy-discogs-effnet-1.json : for json files, you may need to right-click on the link > Save link target as...
    discogs-effnet-bs64-1.pb

  • Create a file called script.py, and paste this code inside it. On line 8, replace audio.wav with the name and extension of your audio file, and save your file

    script.py 
    import json

with open('mood_happy-discogs-effnet-1.json', 'r') as json_file:
    metadata = json.load(json_file)

from essentia.standard import MonoLoader, TensorflowPredictEffnetDiscogs, TensorflowPredict2D

audio = MonoLoader(filename="audio.wav", sampleRate=16000, resampleQuality=4)()
embedding_model = TensorflowPredictEffnetDiscogs(graphFilename="discogs-effnet-bs64-1.pb", output="PartitionedCall:1")
embeddings = embedding_model(audio)

model = TensorflowPredict2D(graphFilename="mood_happy-discogs-effnet-1.pb", output="model/Softmax")
predictions = model(embeddings)

for label, probability in zip(metadata['classes'], predictions.mean(axis=0)):
    print(f'{label}: {probability}')

  • Put these files in your working folder

  • Open PowerShell and run this command

    py script.py
    Result 
    PS E:\Test> py script.py
2024-08-24 19:26:21.417805: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
[   INFO   ] MusicExtractorSVM: no classifier models were configured by default
2024-08-24 19:26:22.285506: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: SSE SSE2 SSE3 SSE4.1 SSE4.2 AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
[   INFO   ] TensorflowPredict: Successfully loaded graph file: `discogs-effnet-bs64-1.pb`
2024-08-24 19:26:23.779818: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:388] MLIR V1 optimization pass is not enabled
[   INFO   ] TensorflowPredict: Successfully loaded graph file: `mood_happy-discogs-effnet-1.pb`
[   INFO   ] TensorflowPredict: Successfully loaded graph file: `mood_happy-discogs-effnet-1.pb`
happy: 0.6639894843101501
non_happy: 0.33601048588752747
PS E:\Test>

Find out more

Clone this wiki locally