Datasets:
license: cc-by-nc-2.0
task_categories:
- text-to-speech
- text-to-audio
- text-to-video
language:
- zh
size_categories:
- 10B<n<100B
Dataset Description
This dataset was recorded by a professional Chinese voice artist, real person, female, the total length is around 40 hours. There are different emotion styles in this dataset, including but not limited to Sad, Calm, Advertising, E-learning. For more details, please refer to the link: www.vodataset.com or email [email protected].
Authorization
The voice artist can sign an authorization with you or your company. The authorization can be of any form, such as paper, voice or video.
Custom Dataset
Yes, the artist can record a new dataset with your scripts, and meet your requirements for the dataset.
Audio Format
48,000Hz, 24bit, wav, mono.
Recording Environment
Professional Recording Studio.
Recording Content
Different texts with different emotion styles: Sad, Calm, Advertising, E-learning
Speakers
1 Professional Female Chinese voice artist.
Language
Chinese Mandarin.
Annotation
Chinese Characters and Pinyin (carefully reviewed by phoneticians).
Useage
TTS/Speech Synthesis.
Licensing Information
Commercial License