Great to see the performance calibration / post processing is now availible.
I’we encountered an error trying to use it on my dataset, i’m suspiscious it might be caused by a high sampling rate of 100KHz.
The performance calibration output yields following message, here shown with 30min audio generated, 10min results in similar error :
Creating job... OK (ID: 4430406)
Generating features for Spectrogram...
Not generating new features: features already generated and no options or files have changed.
No new features, skipping...
Generating features for Spectrogram OK
Running performance calibration on synthetic data...
Copying features from processing blocks...
Copying features from DSP block...
Copying features from DSP block OK
Copying features from processing blocks OK
Copying features from processing blocks...
Copying features from DSP block...
Scheduling job in cluster...
Job started
Impulse details:
Sample rate 100000 Hz
Window size 100 ms
Learn block latency 1 ms
DSP block latency unavailable
Total latency 1 ms
Using minimum latency 25 ms
Match tolerance 500 ms
Target test sample length is 30 minutes, including 5405 test data items.
Warning: Your test data only has 4554 items, so the test sample will only be 25.2747 minutes. Add 851 test data items for a full-length test sample.
Attached to job 4430406...
Traceback (most recent call last):
File "/home/testing.py", line 10, in <module>
ei_testing.testing.run_application_testing(dir_path, options)
File "/app/./resources/libraries/ei_testing/testing.py", line 81, in run_application_testing
sample_path, ground_truth = create_sample(
File "/app/./resources/libraries/ei_testing/testing.py", line 266, in create_sample
sample = sample.overlay(window_audio, position=region.insertion_point)
File "/app/perfcal/.venv/lib/python3.8/site-packages/pydub/audio_segment.py", line 1212, in overlay
seg1, seg2 = AudioSegment._sync(self, seg)
File "/app/perfcal/.venv/lib/python3.8/site-packages/pydub/audio_segment.py", line 440, in _sync
return tuple(
File "/app/perfcal/.venv/lib/python3.8/site-packages/pydub/audio_segment.py", line 441, in <genexpr>
seg.set_channels(channels).set_frame_rate(frame_rate).set_sample_width(sample_width)
File "/app/perfcal/.venv/lib/python3.8/site-packages/pydub/audio_segment.py", line 1017, in set_frame_rate
converted, _ = audioop.ratecv(self._data, self.sample_width,
TypeError: integer argument expected, got float
Application exited with code 1