Dataset · langswap
Studio emotional dialogues in Russian
langswap/dialogs-ru-emotional-conversations
20.6 hours of studio recordings of live theatrical actor dialogues for emotional TTS and ASR.
What's inside
An audio-text corpus in Russian with a total volume of 20.6 hours (less than 1 thousand records). Unlike singly read books or web scraping, professional theatrical actors live-acted out dialogue scenarios in the studio, preserving natural pauses, intonations, and turn-taking transitions.
What you can do with it
- Training and fine-tuning speech synthesis (TTS) models with natural emotions
- Development of conversational voice assistants and automatic speech recognition (ASR) systems
- Tasks
- Text → speech, Automatic speech recognition, Audio classification
- Languages
- Russian
- Size
- up to 1K rows
- Downloads this month
- 5.6K
- Likes
- 27
Models trained on it
How to load it
from datasets import load_dataset
ds = load_dataset("langswap/dialogs-ru-emotional-conversations")