espnet2.speechlm.bin.prepare_audio_lhotse.parse_segments_file

source

espnet2.speechlm.bin.prepare_audio_lhotse.parse_segments_file(segments_path: str, recording_dict: Dict[str, Recording]) → List[Tuple[str, Recording, float, float]]

Parse and validate segments file.

• Parameters:
  • segments_path – Path to segments file
  • recording_dict – Dictionary mapping recording IDs to Recording objects
• Returns: List of validated (segment_id, recording, start, duration) tuples