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 Casual Conversation V2 audio scenario #3158

Merged
merged 6 commits into from
Nov 15, 2024

Conversation

ImKeTT
Copy link
Collaborator

@ImKeTT ImKeTT commented Nov 14, 2024

  1. Fix scenario subset name errors in the audio_run_specs.py
  2. Add missing tags to multilinguality scenarios
  3. Add the Casual Conversation V2 scenario

The Casual Conversation V2 dataset is quite large, with 80 .zip files totaling 30-60GB per file. We'll need to download and unzip it manually into the folder.

For the age and gender classification tasks, I adjusted them to a multiple-choice format.

  • For age classification, while most speakers have specific ages, it’s difficult for current audioLLMs to recognize them accurately. So I grouped the ages into four ranges: "18-30", "31-50", "51+", and "others."
  • For gender classification, which originally had six categories, I simplified it into three options: "male", "female", and "others."

I'm attaching the scenario_state.json files

scenario_state_ccv2_age.json
scenario_state_ccv2_gender.json

@ImKeTT ImKeTT requested a review from teetone November 14, 2024 00:09
@teetone teetone merged commit dd8a58a into stanford-crfm:main Nov 15, 2024
8 checks passed
@ImKeTT ImKeTT deleted the fairness_audio_scenarios branch December 2, 2024 18:56
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.

2 participants