Batch Audio Processing

[systemdevart]

Thank you for this fantastic speech enhancement model!

Currently, it seems that the .process method is only capable of handling a single audio file at a time, which can be quite inefficient for processing large datasets. Are there any existing implementations, plans, or guidelines to support batch audio processing in the model? This feature would significantly improve the usability for large-scale datasets.

[alibabasglab]

Thank you for your interests and suggestions.

You are right that a batch processing capability is more efficient for large datasets. We will consider this option and make our internal check on this.