Upload 6 files

chat_template.jinja

@@ -0,0 +1,45 @@
+
+{%- for message in messages %}
+    {#-- Validate role is a stringified integer --#}
+    {%- if not message['role'] is string or not message['role'].isdigit() %}
+        {{- raise_exception("The role must be an integer or a stringified integer (e.g. '0') designating the speaker id") }}
+    {%- endif %}
+
+    {#-- Validate content is a list --#}
+    {%- set content = message['content'] %}
+    {%- if content is not iterable or content is string %}
+        {{- raise_exception("The content must be a list") }}
+    {%- endif %}
+
+    {#-- Collect content types --#}
+    {%- set content_types = content | map(attribute='type') | list %}
+    {%- set is_last = loop.last %}
+
+    {#-- Last message validation --#}
+    {%- if is_last %}
+        {%- if 'text' not in content_types %}
+            {{- raise_exception("The last message must include one item of type 'text'") }}
+        {%- elif (content_types | select('equalto', 'text') | list | length > 1) or (content_types | select('equalto', 'audio') | list | length > 1) %}
+            {{- raise_exception("At most two items are allowed in the last message: one 'text' and one 'audio'") }}
+        {%- endif %}
+
+    {#-- All other messages validation --#}
+    {%- else %}
+        {%- if content_types | select('equalto', 'text') | list | length != 1
+              or content_types | select('equalto', 'audio') | list | length != 1 %}
+            {{- raise_exception("Each message (except the last) must contain exactly one 'text' and one 'audio' item") }}
+        {%- elif content_types | reject('in', ['text', 'audio']) | list | length > 0 %}
+            {{- raise_exception("Only 'text' and 'audio' types are allowed in content") }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+
+{%- for message in messages %}
+    {{- bos_token }}
+    {{- '[' + message['role'] + ']' }}
+    {{- message['content'][0]['text'] }}
+    {{- eos_token }}
+    {%- if message['content']|length > 1 %}
+        {{- '<|AUDIO|><|audio_eos|>' }}
+    {%- endif %}
+{%- endfor %}

config.json

@@ -0,0 +1,126 @@
+{
+  "architectures": [
+    "CsmForConditionalGeneration"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "audio_eos_token_id": 128003,
+  "audio_num_codebooks": 32,
+  "audio_token_id": 128002,
+  "audio_vocab_size": 2051,
+  "backbone_flavor": "llama-1B",
+  "bos_token_id": 128000,
+  "codebook_eos_token_id": 0,
+  "codebook_pad_token_id": 2050,
+  "codec_config": {
+    "_frame_rate": 12.5,
+    "_name_or_path": "kyutai/mimi",
+    "architectures": [
+      "MimiModel"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "audio_channels": 1,
+    "codebook_dim": 256,
+    "codebook_size": 2048,
+    "compress": 2,
+    "dilation_growth_rate": 2,
+    "head_dim": 64,
+    "hidden_act": "gelu",
+    "hidden_size": 512,
+    "initializer_range": 0.02,
+    "intermediate_size": 2048,
+    "kernel_size": 7,
+    "last_kernel_size": 3,
+    "layer_scale_initial_scale": 0.01,
+    "max_position_embeddings": 8000,
+    "model_type": "mimi",
+    "norm_eps": 1e-05,
+    "normalize": false,
+    "num_attention_heads": 8,
+    "num_filters": 64,
+    "num_hidden_layers": 8,
+    "num_key_value_heads": 8,
+    "num_quantizers": 32,
+    "num_residual_layers": 1,
+    "num_semantic_quantizers": 1,
+    "pad_mode": "constant",
+    "residual_kernel_size": 3,
+    "rope_theta": 10000.0,
+    "sampling_rate": 24000,
+    "sliding_window": 250,
+    "torch_dtype": "float32",
+    "trim_right_ratio": 1.0,
+    "upsample_groups": 512,
+    "upsampling_ratios": [
+      8,
+      6,
+      5,
+      4
+    ],
+    "use_cache": false,
+    "use_causal_conv": true,
+    "use_conv_shortcut": false,
+    "use_streaming": false,
+    "vector_quantization_hidden_dimension": 256
+  },
+  "decoder_flavor": "llama-100M",
+  "depth_decoder_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "backbone_hidden_size": 2048,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 1024,
+    "initializer_range": 0.02,
+    "intermediate_size": 8192,
+    "max_position_embeddings": 33,
+    "mlp_bias": false,
+    "model_type": "csm_depth_decoder_model",
+    "num_attention_heads": 8,
+    "num_codebooks": 32,
+    "num_hidden_layers": 4,
+    "num_key_value_heads": 2,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": {
+      "factor": 32.0,
+      "high_freq_factor": 0.0078125,
+      "low_freq_factor": 0.001953125,
+      "original_max_position_embeddings": 16,
+      "rope_type": "llama3"
+    },
+    "rope_theta": 500000,
+    "torch_dtype": "float32",
+    "use_cache": true,
+    "vocab_size": 2051
+  },
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 2048,
+  "mlp_bias": false,
+  "model_type": "csm",
+  "num_attention_heads": 32,
+  "num_codebooks": 32,
+  "num_hidden_layers": 16,
+  "num_key_value_heads": 8,
+  "pad_token_id": 128002,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 32.0,
+    "high_freq_factor": 0.5,
+    "low_freq_factor": 0.125,
+    "original_max_position_embeddings": 1024,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000,
+  "text_vocab_size": 128256,
+  "tie_codebooks_embeddings": true,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.55.4",
+  "use_cache": true,
+  "vocab_size": 2051
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "depth_decoder_do_sample": true,
+  "depth_decoder_max_new_tokens": 31,
+  "depth_decoder_min_new_tokens": 31,
+  "depth_decoder_temperature": 0.9,
+  "do_sample": true,
+  "max_new_tokens": 125,
+  "pad_token_id": 128002,
+  "temperature": 0.9,
+  "transformers_version": "4.55.4"
+}

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f333a76774a382c3c2d2a602a6fe8a32c9097f8bff46c566e6aac2686230b65
+size 1635765924

model.safetensors.index.json

@@ -0,0 +1,545 @@
+{
+  "metadata": {
+    "total_parameters": 1632100161,
+    "total_size": 6595771780
+  },
+  "weight_map": {
+    "backbone_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone_model.norm.weight": "model-00001-of-00002.safetensors",
+    "codec_model.decoder.layers.0.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.0.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.11.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.11.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.12.block.1.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.12.block.1.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.12.block.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.12.block.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.14.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.14.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.2.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.2.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.3.block.1.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.3.block.1.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.3.block.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.3.block.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.5.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.5.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.6.block.1.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.6.block.1.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.6.block.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.6.block.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.8.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.8.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.9.block.1.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.9.block.1.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.9.block.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder.layers.9.block.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.0.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.1.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.2.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.3.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.4.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.5.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.6.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.decoder_transformer.layers.7.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.downsample.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.0.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.0.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.1.block.1.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.1.block.1.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.1.block.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.1.block.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.10.block.1.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.10.block.1.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.10.block.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.10.block.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.12.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.12.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.14.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.14.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.4.block.1.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.4.block.1.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.4.block.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.4.block.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.6.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.6.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.7.block.1.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.7.block.1.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.7.block.3.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.7.block.3.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.9.conv.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder.layers.9.conv.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.0.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.1.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.2.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.3.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.4.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.5.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.6.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.input_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.mlp_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.encoder_transformer.layers.7.self_attn_layer_scale.scale": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.input_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.0.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.0.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.0.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.1.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.1.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.1.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.10.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.10.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.10.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.11.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.11.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.11.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.12.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.12.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.12.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.13.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.13.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.13.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.14.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.14.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.14.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.15.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.15.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.15.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.16.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.16.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.16.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.17.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.17.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.17.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.18.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.18.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.18.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.19.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.19.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.19.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.2.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.2.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.2.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.20.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.20.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.20.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.21.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.21.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.21.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.22.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.22.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.22.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.23.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.23.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.23.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.24.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.24.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.24.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.25.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.25.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.25.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.26.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.26.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.26.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.27.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.27.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.27.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.28.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.28.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.28.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.29.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.29.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.29.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.3.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.3.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.3.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.30.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.30.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.30.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.4.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.4.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.4.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.5.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.5.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.5.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.6.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.6.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.6.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.7.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.7.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.7.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.8.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.8.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.8.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.9.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.9.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.layers.9.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.acoustic_residual_vector_quantizer.output_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.semantic_residual_vector_quantizer.input_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.semantic_residual_vector_quantizer.layers.0.codebook.cluster_usage": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.semantic_residual_vector_quantizer.layers.0.codebook.embed_sum": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.semantic_residual_vector_quantizer.layers.0.codebook.initialized": "model-00002-of-00002.safetensors",
+    "codec_model.quantizer.semantic_residual_vector_quantizer.output_proj.weight": "model-00002-of-00002.safetensors",
+    "codec_model.upsample.conv.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.codebooks_head.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.embed_tokens.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.inputs_embeds_projector.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "depth_decoder.model.norm.weight": "model-00002-of-00002.safetensors",
+    "embed_text_tokens.weight": "model-00001-of-00002.safetensors",
+    "lm_head.weight": "model-00001-of-00002.safetensors"
+  }
+}

preprocessor_config.json

@@ -0,0 +1,11 @@
+{
+  "chunk_length_s": null,
+  "feature_extractor_type": "EncodecFeatureExtractor",
+  "feature_size": 1,
+  "overlap": null,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "CsmProcessor",
+  "return_attention_mask": true,
+  "sampling_rate": 24000
+}