moraMora / Datasets / "speech recognition" / MLCommons/unsupervised_peoples_speech
Unsupervised Peoples SpeechFree and public
The Unsupervised Peoples Speech Dataset is a compilation of audiofiles extracted from Archive.org that is licensed for academic and commercial usage under CC-BY and CC-BY-SA licenses. It includes more than one million hours of audio with a diverse set of speakers.
- Published on
Hugging FaceMLCommons/unsupervised_peoples_speech- Price
- Free
- License
- not stated
- Allows
- Unknown: read the license before you use it
- Made for
- automatic-speech-recognition, audio-classification
- Languages
- eng
- Downloads
- 15,406
- Last updated
- 2025-02-27
Mora did not check this dataset. "Allows" reads the declared license only.