This dataset is made up of real-world audio recordings from different locations in Hangzhou, China, aimed at enhancing the performance of Speech-to-Text models. It is encrypted for privacy protection. Researchers who wish to gain access must reach out to the uploader with a detailed research plan and agree to confidentiality requirements.
- Downloads last month
- 5