azalea/SpeechGenderAnalysis
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[O] Comment

Browse Source
This commit is contained in:
Hykilpikonna
2022-03-21 02:52:23 -04:00
parent e592d1a15b
commit 21b6de3831
1 changed files with 5 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/formant.py
+5 -2
View File
@@ -365,16 +365,19 @@ if __name__ == '__main__':
############ ############
# 1. Compute and save all the frequency (pitch, f0, f1, f2) for vox1 # 1. Compute and save all the frequency (pitch, f0, f1, f2) for vox1
# For each audio, a file <audio-name>.npy will be saved, with each row representing 10ms data
# compute_audio_vox_celeb(compute_audio_freq) # compute_audio_vox_celeb(compute_audio_freq)
# 2. Combine statistics for each person in vox1 # 2. Combine and save statistics for each person in vox1
call_id_vox_celeb(combine_id_freq) # For each person, stats.json will be saved, containing statistics of all of their audios
# call_id_vox_celeb(combine_id_freq)
# 3. Collect statistics and draw visualizations # 3. Collect statistics and draw visualizations
collect_visualize_freq() collect_visualize_freq()
########### ###########
# 1. Compute and save all the spectral tilt for vox1 # 1. Compute and save all the spectral tilt for vox1
# For each audio, a file <audio-name>.json will be saved with tilt value in it
# compute_audio_vox_celeb(compute_audio_tilt) # compute_audio_vox_celeb(compute_audio_tilt)
# 2. Combine statistics for each person in vox1 # 2. Combine statistics for each person in vox1