Inquiry About the License of Pretrained Models

[Apple-523]

Dear PaddleSpeech Team,

First of all, thank you for open-sourcing such an excellent and comprehensive speech processing framework along with pretrained models. Your work greatly contributes to the advancement and application of speech technologies.

I am currently considering using the pretrained models from PaddleSpeech in a commercial project. To ensure compliance, I would like to kindly ask about the licensing of the pretrained model weight files:

• Are all officially released pretrained model weights licensed under the Apache 2.0 License?
• Do all the pretrained models explicitly include license declarations or LICENSE files?
• Are the datasets used to train these models all cleared for commercial use? If there are restrictions, could you please recommend datasets and corresponding models that are suitable for commercial deployment?

Thank you very much for your time and support. I look forward to your reply and wish the PaddleSpeech project continued success!

Best regards,
Apple-523

[zxcd]

All we released pretrained models follow Apache 2.0 License, also we use open-source datasets for training.

For more detail, you can check our readme to see what kind of datasets is used for training each model, such as README.md.

[Apple-523]

Thinks！