name: "preprocessing"

data_base_dir: ???

defaults:
  - trim: energy

config:
  _target_: scripts.dataset_processing.tts.audio_processing.preprocess_audio.AudioPreprocessingConfig
  input_manifest: ${data_base_dir}/manifest.json
  output_manifest: ${data_base_dir}/manifest_processed.json
  output_dir: ${data_base_dir}/audio_processed
  num_workers: -1
  max_entries: 0
  output_sample_rate: 0
  volume_level: 0.95
  min_duration: 0.5
  max_duration: 10.0
  filter_file: ${data_base_dir}/filtered_utts.json