in src/diarizers/data/synthetic_pipeline.py [0:0]
def __init__(
self,
dataset_name: str,
subset: str,
split: str,
speaker_column_name: str,
audio_column_name: str,
min_samples_per_speaker: int = 10,
nb_speakers_from_dataset: int = -1,
sample_rate: int = 16000,
num_meetings: int = 1600,
nb_speakers_per_meeting: int = 3,
segments_per_meeting: int = 16,
normalize: bool = True,
augment: bool = False,
overlap_proba: float = 0.3,
overlap_length: float = 3,
random_gain: bool = False,
add_silence: bool = True,
silence_duration: float = 3,
silence_proba: float = 0.7,
denoise: bool = False,
bn_path: str = None,
ir_path: str = None,
num_proc: int = 2,