espnet2.speechlm.bin.prepare_audio_lhotse.prepare_audio_lhotse

source

espnet2.speechlm.bin.prepare_audio_lhotse.prepare_audio_lhotse(wav_scp: str, segments: str | None, output_dir: str, num_jobs: int, log_level: str)

Process Kaldi wav.scp and segments to create Lhotse manifests.

• Parameters:
  • wav_scp – Path to Kaldi wav.scp file
  • segments – Path to Kaldi segments file (optional)
  • output_dir – Directory to save Lhotse manifest files
  • num_jobs – Number of parallel jobs for processing
  • log_level – Logging level