Skip to content

Latest commit

 

History

History
28 lines (23 loc) · 1.08 KB

File metadata and controls

28 lines (23 loc) · 1.08 KB

1722-Hours-Near-field-Conference-Speech-Data

Description

Mandarin(China) Near-field Conference speech dataset, collected the output by AU central console mixer in real speech scenes. It has a natural pronunciation without environmental noise almost, covers a variety of topics. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.

For more details, please refer to the link: https://www.nexdata.ai/datasets/1066?source=Github

Specifications

Format

44.1kHz, 16bit, wav, dual channel.

Recording environment

Mixed

Recording content

lectures on science and technology, training, publicity, etc.

Device

AU Center Console Mixer

Country

China(CHN)

Language

Mandarin

Features of annotation

annotating for the transcription text, speaker identification and gender

Accuracy Rate

Sentence Accuracy Rate(SAR) 97%

Licensing Information

Commercial License