---
task_categories:
- automatic-speech-recognition
language:
- ro
pretty_name: Echo
---

# Echo

Echo is a speech dataset for Romanian language crowd-sourced from the community.
The dataset contains over 300 hours of speech data from 300 speakers and is
available for non-commercial research purposes. The dataset is collected using
the [Echo platform](https://echo.readerbench.com/).