name: "ClusterDiarizer"

num_workers: 1
sample_rate: 16000
batch_size: 64
device: cpu
verbose: True

diarizer:
  manifest_filepath: .temp/manifest.json
  out_dir: .temp
  oracle_vad: False
  collar: 0.25
  ignore_overlap: True

  vad:
    model_path: vad_multilingual_marblenet
    external_vad_manifest: null
    parameters:
      window_length_in_sec: 0.15
      shift_length_in_sec: 0.01
      smoothing: "median"
      overlap: 0.5
      onset: 0.1
      offset: 0.1
      pad_onset: 0.1
      pad_offset: 0
      min_duration_on: 0
      min_duration_off: 0.2
      filter_speech_first: True

  speaker_embeddings:
    model_path: titanet_large
    parameters:
      window_length_in_sec: [ 1.5,1.25,1.0,0.75,0.5 ]
      shift_length_in_sec: [ 0.75,0.625,0.5,0.375,0.25 ]
      multiscale_weights: [ 1,1,1,1,1 ]
      save_embeddings: True

  clustering:
    parameters:
      oracle_num_speakers: False
      max_num_speakers: 8
      enhanced_count_thres: 80
      max_rp_threshold: 0.25
      sparse_search_volume: 30
      maj_vote_spk_count: False
      chunk_cluster_count: 50
      embeddings_per_chunk: 10000

  msdd_model:
    model_path: diar_msdd_telephonic
    parameters:
      use_speaker_model_from_ckpt: True
      infer_batch_size: 25
      sigmoid_threshold: [ 0.7 ]
      seq_eval_mode: False
      split_infer: True
      diar_window_length: 50
      overlap_infer_spk_limit: 5

  asr:
    model_path: stt_en_conformer_ctc_large
    parameters:
      asr_based_vad: False
      asr_based_vad_threshold: 1.0
      asr_batch_size: null
      decoder_delay_in_sec: null
      word_ts_anchor_offset: null
      word_ts_anchor_pos: "start"
      fix_word_ts_with_VAD: False
      colored_text: False
      print_time: True
      break_lines: False

    ctc_decoder_parameters:
      pretrained_language_model: null
      beam_width: 32
      alpha: 0.5
      beta: 2.5

    realigning_lm_parameters:
      arpa_language_model: null
      min_number_of_words: 3
      max_number_of_words: 10
      logprob_diff_threshold: 1.2