Depression speech analysis on E-DAIC

Speech analysis using the extended Distress Analysis Interview Corpus. Uses Librosa (version 0.9.2) for audio analysis.

Sign and submit a form to download the speech dataset (see link above)
Segment the speech files with slicedAudio.py. For that you need the transcription csv files from the dataset. Dont change the speech file names!
After segmentation place the audio files into the interviews/-folder
Run cells from scatter_audio_feat_depress.ipynb

License

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
.ipynb_checkpoints		.ipynb_checkpoints
audio_test_files		audio_test_files
extracted_feature_results		extracted_feature_results
participants_info		participants_info
plots		plots
vosk		vosk
.gitignore		.gitignore
README.md		README.md
my-voice-analysis.ipynb		my-voice-analysis.ipynb
noise reduction.ipynb		noise reduction.ipynb
participant statistics.ipynb		participant statistics.ipynb
scatter_audio_feat_depress.ipynb		scatter_audio_feat_depress.ipynb
slicedAudio.py		slicedAudio.py
spectrum plotting.ipynb		spectrum plotting.ipynb