Baseline

Python Modules

recipes.icassp_2023.baseline.enhance

Run the dummy enhancement.

recipes.icassp_2023.baseline.evaluate

Evaluate the enhanced signals using a combination of HASPI and HASQI metrics

recipes.icassp_2023.baseline.report_score

Run the dummy enhancement.

Configuration Files

config.yaml

 1path:
 2  root: ./ # root folder of the project
 3  metadata_dir: ${path.root}/clarity_data/metadata
 4  scenes_listeners_file: ${path.metadata_dir}/scenes_listeners.dev.json
 5  listeners_file: ${path.metadata_dir}/listeners.json
 6  scenes_folder: ${path.root}/clarity_data/dev/scenes
 7  exp_folder: ./exp # folder to store enhanced signals and final results
 8
 9nalr:
10  nfir: 220
11  sample_rate: 44100
12
13compressor:
14  threshold: 0.35
15  attenuation: 0.1
16  attack: 50
17  release: 1000
18  rms_buffer_size: 0.064
19
20soft_clip: True
21
22evaluate:
23  set_random_seed: True
24  small_test: False
25
26# hydra config
27hydra:
28  run:
29    dir: ${path.exp_folder}
30  job:
31    chdir: True