Skip to content

Latest commit

 

History

History
12 lines (7 loc) · 526 Bytes

README.md

File metadata and controls

12 lines (7 loc) · 526 Bytes

highlow_release

Supplementary information for the Interspeech 2024 paper:

Liang et al. 2024. Improving Audio Classification with Low-Sampled Microphone Input: An Empirical Study Using Model Self-Distillation.

======

The details of the hyper-parameters we used: Supplementary_information-camera_ready.pdf

The main Python scripts: /scripts/esc_folds.py, /scripts/tau2019_folds.py (to pre-process audio samples), /scripts/kd.py (self-distillation)

Demonstration of the re-sampling outcomes: /librispeech_effects/