Closed StevenJBorik closed 1 year ago
Can you check if this fixes your problem? https://github.com/urinieto/msaf/issues/120#issuecomment-1469284644
Thanks for the quick response Urinieto! I'll try this later tonight and let you know.
Hey Urinieto, I'm going to go through the rest of my other requirements versions tomorrow, but after going from librosa 0.10.1 to 0.6.1:
pip3 show librosa Name: librosa Version: 0.6.1
I get a this error that I need to investigate more, both in this test script and the same in my API using this.
python3 msafTest.py
Traceback (most recent call last):
File "/Users/stev/Dev/msafTest.py", line 3, in
Please, consider downgrading to Python 3.9. Python 3.11 has been giving me some problems of compatibility with some scientific packages.
Gotcha, I will try this as well and keep this thread posted. Thanks again!
Hey Urinieto, downgrading to python 3.9 and keeping librosa at 0.10 seems to be the trick for me. Moving librosa down to 0.6.1 still gives me the same error but with .10 and 3.9 I'm not getting any errors after multiple tests, I'm going to close the ticket. Thanks so much for your help, this library is awesome. Working on an exciting project and this is essential.
So happy to hear! Thanks for reporting :)
Hi, I've been using this for a couple months within a Python API but as of recently when I try to analyze a new audio file I get the following error:
MSAF error: Computation of the features is needed for current parameters but no audio file was found.Please, change your parameters or add the audio file in /location/file.mp3.
The file is indeed in the location, playable, and not corrupted. The behavior I'm noticing is that my features_msaf_tmp.json only has 1 set of features for a song in it, and when I upload that song, it processes it correctly, but when I add a new song I receive the title error, and the json file still has the data for the previous song. If I try to remove the json file in hopes of the library generating a new one for the new song, I expectedly get:
Traceback (most recent call last): File "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/msaf/base.py", line 220, in read_features with open(self.file_struct.features_file) as f: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ FileNotFoundError: [Errno 2] No such file or directory: '.features_msaf_tmp.json'
Next, I attempted to reproduce from scratch the using the starter code below, but still get the same error as no features_msaf_temp.json is generated.
It's not immediately clear to me how I should approach debugging this from the documentation. Any help would be immensely appreciated. Below is the starter example I reference earlier.
from future import print_function import msaf
audio_file = "Very Ape.mp3"
boundaries, labels = msaf.process(audio_file) print('Estimated boundaries:', boundaries)
out_file = 'segments.txt' print('Saving output to %s' % out_file) msaf.io.write_mirex(boundaries, labels, out_file)
try: evals = msaf.eval.process(audio_file) print(evals) except msaf.exceptions.NoReferencesError: file_struct = msaf.input_output.FileStruct(audio_file) print("No references found in {}. No evaluation performed.".format( file_struct.ref_file))