Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add VAD to Supervisions in LibriLight Recipe #1280

Merged
merged 3 commits into from
Feb 3, 2024

Conversation

yfyeung
Copy link
Contributor

@yfyeung yfyeung commented Feb 2, 2024

This PR introduces VAD information to the LibriLight recipe when creating supervisions.

@yfyeung yfyeung changed the title Add vad info to supervisions Add VAD to Supervisions in LibriLight Recipe Feb 2, 2024
@yfyeung
Copy link
Contributor Author

yfyeung commented Feb 2, 2024

  • large
Cut statistics:
╒═══════════════════════════╤═════════════╕
│ Cuts count:               │ 16651603    │
├───────────────────────────┼─────────────┤
│ Total duration (hh:mm:ss) │ 47940:10:19 │
├───────────────────────────┼─────────────┤
│ mean                      │ 10.4        │
├───────────────────────────┼─────────────┤
│ std                       │ 16.2        │
├───────────────────────────┼─────────────┤
│ min                       │ 0.5         │
├───────────────────────────┼─────────────┤
│ 25%                       │ 3.7         │
├───────────────────────────┼─────────────┤
│ 50%                       │ 6.8         │
├───────────────────────────┼─────────────┤
│ 75%                       │ 12.2        │
├───────────────────────────┼─────────────┤
│ 99%                       │ 61.0        │
├───────────────────────────┼─────────────┤
│ 99.5%                     │ 82.2        │
├───────────────────────────┼─────────────┤
│ 99.9%                     │ 165.1       │
├───────────────────────────┼─────────────┤
│ max                       │ 4151.8      │
├───────────────────────────┼─────────────┤
│ Recordings available:     │ 16651603    │
├───────────────────────────┼─────────────┤
│ Features available:       │ 0           │
├───────────────────────────┼─────────────┤
│ Supervisions available:   │ 16651603    │
╘═══════════════════════════╧═════════════╛
Speech duration statistics:
╒══════════════════════════════╤═════════════╤══════════════════════╕
│ Total speech duration        │ 47940:10:19 │ 100.00% of recording │
├──────────────────────────────┼─────────────┼──────────────────────┤
│ Total speaking time duration │ 47940:10:19 │ 100.00% of recording │
├──────────────────────────────┼─────────────┼──────────────────────┤
│ Total silence duration       │ 00:00:00    │ 0.00% of recording   │
╘══════════════════════════════╧═════════════╧══════════════════════╛
  • medium
Cut statistics:
╒═══════════════════════════╤════════════╕
│ Cuts count:               │ 1778314    │
├───────────────────────────┼────────────┤
│ Total duration (hh:mm:ss) │ 4747:27:03 │
├───────────────────────────┼────────────┤
│ mean                      │ 9.6        │
├───────────────────────────┼────────────┤
│ std                       │ 14.4       │
├───────────────────────────┼────────────┤
│ min                       │ 0.5        │
├───────────────────────────┼────────────┤
│ 25%                       │ 3.5        │
├───────────────────────────┼────────────┤
│ 50%                       │ 6.5        │
├───────────────────────────┼────────────┤
│ 75%                       │ 11.5       │
├───────────────────────────┼────────────┤
│ 99%                       │ 53.8       │
├───────────────────────────┼────────────┤
│ 99.5%                     │ 71.4       │
├───────────────────────────┼────────────┤
│ 99.9%                     │ 137.9      │
├───────────────────────────┼────────────┤
│ max                       │ 2661.8     │
├───────────────────────────┼────────────┤
│ Recordings available:     │ 1778314    │
├───────────────────────────┼────────────┤
│ Features available:       │ 0          │
├───────────────────────────┼────────────┤
│ Supervisions available:   │ 1778314    │
╘═══════════════════════════╧════════════╛
Speech duration statistics:
╒══════════════════════════════╤════════════╤══════════════════════╕
│ Total speech duration        │ 4747:27:03 │ 100.00% of recording │
├──────────────────────────────┼────────────┼──────────────────────┤
│ Total speaking time duration │ 4747:27:03 │ 100.00% of recording │
├──────────────────────────────┼────────────┼──────────────────────┤
│ Total silence duration       │ 00:00:00   │ 0.00% of recording   │
╘══════════════════════════════╧════════════╧══════════════════════╛
  • small
Cut statistics:
╒═══════════════════════════╤═══════════╕
│ Cuts count:               │ 200009    │
├───────────────────────────┼───────────┤
│ Total duration (hh:mm:ss) │ 526:27:49 │
├───────────────────────────┼───────────┤
│ mean                      │ 9.5       │
├───────────────────────────┼───────────┤
│ std                       │ 13.5      │
├───────────────────────────┼───────────┤
│ min                       │ 0.5       │
├───────────────────────────┼───────────┤
│ 25%                       │ 3.5       │
├───────────────────────────┼───────────┤
│ 50%                       │ 6.6       │
├───────────────────────────┼───────────┤
│ 75%                       │ 11.4      │
├───────────────────────────┼───────────┤
│ 99%                       │ 51.0      │
├───────────────────────────┼───────────┤
│ 99.5%                     │ 69.4      │
├───────────────────────────┼───────────┤
│ 99.9%                     │ 149.1     │
├───────────────────────────┼───────────┤
│ max                       │ 1120.4    │
├───────────────────────────┼───────────┤
│ Recordings available:     │ 200009    │
├───────────────────────────┼───────────┤
│ Features available:       │ 0         │
├───────────────────────────┼───────────┤
│ Supervisions available:   │ 200009    │
╘═══════════════════════════╧═══════════╛
Speech duration statistics:
╒══════════════════════════════╤═══════════╤══════════════════════╕
│ Total speech duration        │ 526:27:49 │ 100.00% of recording │
├──────────────────────────────┼───────────┼──────────────────────┤
│ Total speaking time duration │ 526:27:49 │ 100.00% of recording │
├──────────────────────────────┼───────────┼──────────────────────┤
│ Total silence duration       │ 00:00:00  │ 0.00% of recording   │
╘══════════════════════════════╧═══════════╧══════════════════════╛

id=file_name + "_" + str(segment_seq),
recording_id=file_name,
start=vad_info[0],
duration=vad_info[1] - vad_info[0],
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I’d suggest using lhotse.utils.add_durations() here to avoid annoying rounding errors from subtraction of two floats.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sure.

@pzelasko pzelasko merged commit 00abc09 into lhotse-speech:master Feb 3, 2024
8 of 10 checks passed
@pzelasko pzelasko added this to the v1.21 milestone Feb 3, 2024
@pzelasko
Copy link
Collaborator

pzelasko commented Feb 3, 2024

Thanks!!

@yfyeung yfyeung deleted the librilight_vad branch February 4, 2024 03:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants