Skip to content

Dataset · deepvk

Non-verbal vocalizations for TTS

deepvk/NonverbalTTS

17 hours of English speech with annotations of non-verbal sounds and emotions for expressive TTS.

What's inside

A 17-hour open English speech corpus with aligned text annotations of non-verbal vocalizations and emotional categories. Includes 10 non-verbal types: breathing, laughter, sighs, sneezing, coughing, throat clearing, groans, grunts, snoring, and sniffling. Also features 8 annotated emotional categories; the task is text-to-speech, with metadata size 1K-10K.

What you can do with it

  • Train expressive TTS with non-verbal vocalizations
  • Fine-tune a TTS model on emotions
  • Evaluate laughter and sigh generation in TTS
Tasks
Text → speech
Languages
English
Size
1K–10K rows
Downloads this month
1.9K
Likes
77

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("deepvk/NonverbalTTS")