Datasets:

Loke-60000
/

Christina-TTS-I

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

Dataset Details

Duration: 4h22m9s | 262 minutes
Number of samples: 1749
Average duration of a single sample: ~6 seconds

Update History

October 3rd, 2025: Dataset updated with 554 more samples, increasing total duration from 2h34m58s (1195 samples) to 4h22m9s (1749 samples).

Emotional Tags

laughs - 336, sighs - 156, giggles - 76, chuckles - 20, scoffs - 8, moans - 8,
stutters - 6, whispers - 2, singing - 6, nervous laughter - 6, gasps - 4,
smacks lips - 4, clears throat - 2, coughs - 2, yawning - 2, breathes deeply - 2,
music - 4, fabric rustling - 2, tapping sounds - 2, and various other expressions

Privacy & Ethics

The dataset is created from consented private VA recordings and not Asami Imai.
Raw data not released out of respect for the VA.
Only trained models will be shared.

Acknowledgments

Some parts of the dataset includes a modified version of MrDragonFox/Elise

Todo

More consistant tags
Japanese language
More languages