CNN-based audio segmentation toolkit. Allows to detect speech, music, noise and speaker gender. Has been designed for large scale gender equality studies based on speech time per gender.
OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Ubuntu 18.04
TensorFlow version: 2.3.0
Python version: 3.7.7
Running on GPU or CPU: GPU
CUDA/cuDNN version (if GPU is used): 11.0
Using Docker: No
Expected Behavior
inaspeechsegmenter.py -i https://domain.tld/file.mp3 -o .
should download and process the file and output to ./file.csv.
At some point, it should print
1/1 [('./file.csv', 0, "ok")]
Current Behavior
Instead, it fails with
1/1 [('./file.csv', 2, "error: <class 'AssertionError'>")]
Additional infos
I work behind an HTTP proxy, but it's correctly set in my environment and I can curl the file with no issue.
I have not been able to reproduce this on another computer. I tested on Archlinux with the same configuration and it works fine. (but is not behind a proxy)
After settings my proxy configuration with gsettings instead of HTTP(S)_PROXY env variable, it works fine. I suspect ffmpeg doesn't use the HTTP(S)_PROXY env variable, so not an inaSpeechSegmenter issue.
Not able to process files over HTTP.
System information
Expected Behavior
inaspeechsegmenter.py -i https://domain.tld/file.mp3 -o .
should download and process the file and output to./file.csv
. At some point, it should print1/1 [('./file.csv', 0, "ok")]
Current Behavior
Instead, it fails with
1/1 [('./file.csv', 2, "error: <class 'AssertionError'>")]
Additional infos
I work behind an HTTP proxy, but it's correctly set in my environment and I can
curl
the file with no issue. I have not been able to reproduce this on another computer. I tested on Archlinux with the same configuration and it works fine. (but is not behind a proxy)