config.json

{
  "acostic_vae_dim": 64,
  "acoustic_tokenizer_config": {
    "channels": 1,
    "depths": [
      3,
      3,
      3,
      3,
      3,
      3,
      8
    ],
    "downsampling_ratios": [
      2,
      2,
      4,
      5,
      5,
      8
    ],
    "ffn_expansion": 4,
    "hidden_act": "gelu",
    "hidden_size": 64,
    "initializer_range": 0.01,
    "kernel_size": 7,
    "layer_scale_init_value": 1e-06,
    "model_type": "vibevoice_acoustic_tokenizer",
    "num_filters": 32,
    "rms_norm_eps": 1e-05,
    "vae_std": 0.625,
    "weight_init_value": 0.01
  },
  "architectures": [
    "VibeVoiceForConditionalGeneration"
  ],
  "audio_bos_token_id": 151652,
  "audio_diffusion_token_id": 151654,
  "audio_eos_token_id": 151653,
  "ddpm_beta_schedule": "squaredcos_cap_v2",
  "ddpm_num_inference_steps": 20,
  "ddpm_num_steps": 1000,
  "dtype": "bfloat16",
  "eos_token_id": 151643,
  "frequency_embedding_size": 256,
  "hidden_act": "silu",
  "intermediate_size": 10752,
  "mlp_bias": false,
  "model_type": "vibevoice",
  "num_diffusion_steps": 10,
  "num_head_layers": 4,
  "pad_token_id": 151643,
  "prediction_type": "v_prediction",
  "rms_norm_eps": 1e-05,
  "semantic_tokenizer_config": {
    "causal": true,
    "channels": 1,
    "conv_bias": true,
    "conv_norm": "none",
    "corpus_normalize": 0.0,
    "disable_last_norm": true,
    "encoder_depths": "3-3-3-3-3-3-8",
    "encoder_n_filters": 32,
    "encoder_ratios": [
      8,
      5,
      5,
      4,
      2,
      2
    ],
    "fix_std": 0,
    "layer_scale_init_value": 1e-06,
    "layernorm": "RMSNorm",
    "layernorm_elementwise_affine": true,
    "layernorm_eps": 1e-05,
    "mixer_layer": "depthwise_conv",
    "model_type": "vibevoice_semantic_tokenizer",
    "pad_mode": "constant",
    "std_dist_type": "none",
    "vae_dim": 128,
    "weight_init_value": 0.01
  },
  "semantic_tokenizer_encoder_config": {
    "channels": 1,
    "depths": [
      3,
      3,
      3,
      3,
      3,
      3,
      8
    ],
    "downsampling_ratios": [
      2,
      2,
      4,
      5,
      5,
      8
    ],
    "ffn_expansion": 4,
    "hidden_act": "gelu",
    "hidden_size": 128,
    "initializer_range": 0.01,
    "kernel_size": 7,
    "layer_scale_init_value": 1e-06,
    "model_type": "vibevoice_acoustic_tokenizer_encoder",
    "num_filters": 32,
    "rms_norm_eps": 1e-05,
    "vae_std": 0.625,
    "weight_init_value": 0.01
  },
  "text_config": {
    "attention_dropout": 0.0,
    "bos_token_id": null,
    "dtype": "bfloat16",
    "eos_token_id": null,
    "hidden_act": "silu",
    "hidden_size": 3584,
    "initializer_range": 0.02,
    "intermediate_size": 18944,
    "layer_types": [
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention",
      "full_attention"
    ],
    "max_position_embeddings": 32768,
    "max_window_layers": 28,
    "model_type": "qwen2",
    "num_attention_heads": 28,
    "num_hidden_layers": 28,
    "num_key_value_heads": 4,
    "pad_token_id": null,
    "rms_norm_eps": 1e-06,
    "rope_parameters": {
      "rope_theta": 1000000.0,
      "rope_type": "default"
    },
    "sliding_window": null,
    "tie_word_embeddings": false,
    "use_cache": true,
    "use_mrope": false,
    "use_sliding_window": false,
    "vocab_size": 152064
  },
  "tie_word_embeddings": false,
  "transformers_version": "5.3.0.dev0",
  "vocab_size": 152064
}