add details

audio_tokenizer_config.json

@@ -0,0 +1,5 @@
+{
+  "audio_tokenizer_class": "DacModel",
+  "audio_tokenizer_name_or_path": "descript/dac_44khz"
+}
+

config.json

@@ -22,7 +22,6 @@
     "num_key_value_heads": 4,
     "rope_scaling": null,
     "rope_theta": 10000.0,
-    "use_cache": true,
     "vocab_size": 1028
   },
   "delay_pattern": [
@@ -59,6 +58,6 @@
   "norm_eps": 1e-05,
   "pad_token_id": 1025,
   "torch_dtype": "float32",
-  "transformers_version": "4.53.3",
-  "use_cache": true
+  "transformers_version": "4.53.0.dev0"
 }
+

generation_config.json

@@ -6,6 +6,7 @@
   "max_length": 3072,
   "pad_token_id": 1025,
   "temperature": 1.8,
-  "top_p": 0.9,
-  "transformers_version": "4.53.3"
+  "top_k": 50,
+  "top_p": 0.90,
+  "transformers_version": "4.53.0.dev0"
 }

preprocessor_config.json

@@ -1,8 +1,10 @@
 {
-  "feature_extractor_type": "DacFeatureExtractor",
+  "feature_extractor_type": "DiaFeatureExtractor",
+  "feature_size": 1,
+  "hop_length": 512,
+  "padding_side": "right",
+  "padding_value": 0.0,
   "processor_class": "DiaProcessor",
-  "audio_eos_value": 1024,
-  "audio_pad_value": 1025,
-  "audio_bos_value": 1026
+  "return_attention_mask": true,
+  "sampling_rate": 44100
 }
-

special_token_map.json

@@ -0,0 +1,17 @@
+{
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}
+

tokenizer_config.json

@@ -0,0 +1,37 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[S1]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "2": {
+      "content": "[S2]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "extra_special_tokens": {},
+  "max_length": 1024,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "processor_class": "DiaProcessor",
+  "tokenizer_class": "DiaTokenizer",
+  "unk_token": "<pad>"
+}
+