audiomate

Provides a single API to load multiple audio datasets like ESC-50 and LibriSpeech, as shown in the example code for seamless access and reduced setup time.

Includes built-in functions for validation, splitting, merging, and filtering datasets, detailed in the documentation for robust data handling and preprocessing.

Supports extracting audio features directly within the library, facilitating ML pipeline setup without needing additional external tools for common tasks.

Compatible with formats from Kaldi, DeepSpeech, and others, easing integration with popular ML frameworks and reducing conversion overhead.