100000-hours-English-Unsupervised-speech-dataset

---

Description

English Unsupervised speech dataset, covers dialogues or monologues in 28 common domains, such as tutorials, podcasts, food, entertainment, news, daily vlogs, fashion and style, interviews, photography and photo, skin care and beauty, etc., mirrors real-world interactions, enhancing model performance in real and complex tasks. 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/speechrecog?source=Github

Specifications

Format

16k Hz, 16 bit, wav, mono channel

Data source

public resources

Content category

Dialogue or monologue in 28 common domains, such as tutorials, podcasts, food, entertainment, news, daily vlogs, fashion and style, interviews, photography and photo, skin care and beauty, etc.

Language

English

Country

the United States(USA), the United Kingdom(GBR), Canada(CAN), Australia(AUS), India(IND), Philippine(PHL), etc.

Recording condition

Mixed(indoor, public place, entertainment,etc.)

Language(Region) Code

en-US,en-GB,en-AU,en-CA,en-IN,en-PH,etc.

Licensing Information

Commercial License