{
  "architectures": [
    "AudioSegModel"
  ],
  "auto_map": {
    "AutoConfig": "configuration_audioseg.AudioSegConfig",
    "AutoModel": "modeling_audioseg.AudioSegModel"
  },
  "chunk_size_sec": 6.0,
  "dtype": "float32",
  "emb_dim": 384,
  "encoder_chunk_size_sec": 30.0,
  "encoder_dim": 1280,
  "encoder_frames_per_chunk": 1500,
  "max_frames_per_segment": 512,
  "model_type": "audioseg",
  "roformer_dim_feedforward": 2048,
  "roformer_nhead": 8,
  "roformer_num_layers": 12,
  "sample_rate": 16000,
  "segment_transformer_ff_mult": 4,
  "segment_transformer_heads": 4,
  "segment_transformer_num_layers": 3,
  "threshold": 0.5,
  "transformers_version": "4.57.0.dev0",
  "whisper_model": "openai/whisper-large-v3"
}