Training in progress, step 200

config.json

@@ -8,7 +8,7 @@
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.0,
+  "attention_dropout": 0.05,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
   "codevector_dim": 1024,
@@ -50,16 +50,16 @@
   "feat_extract_activation": "gelu",
   "feat_extract_dropout": 0.0,
   "feat_extract_norm": "layer",
-  "feat_proj_dropout": 0.0,
+  "feat_proj_dropout": 0.05,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.05,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
+  "hidden_dropout": 0.05,
   "hidden_size": 1280,
   "initializer_range": 0.02,
   "intermediate_size": 5120,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.0,
+  "layerdrop": 0.05,
   "mask_feature_length": 10,
   "mask_feature_min_masks": 0,
   "mask_feature_prob": 0.0,
@@ -77,7 +77,7 @@
   "num_hidden_layers": 48,
   "num_negatives": 100,
   "output_hidden_size": 1280,
-  "pad_token_id": 34,
+  "pad_token_id": 0,
   "proj_codevector_dim": 1024,
   "tdnn_dilation": [
     1,
@@ -102,6 +102,6 @@
   ],
   "transformers_version": "4.57.2",
   "use_weighted_layer_sum": false,
-  "vocab_size": 37,
+  "vocab_size": 31,
   "xvector_output_dim": 512
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9492b014ce3d9c18e8bb535633f622efc0e759e19fa093802e6e4c65c4b6e109
-size 3858921668
+oid sha256:5f5ab40a8af65b932e57d869a70a72c897bb521dada361fbcc33f8d2f1a5772d
+size 3858890924

preprocessor_config.json

@@ -3,7 +3,8 @@
   "feature_extractor_type": "Wav2Vec2FeatureExtractor",
   "feature_size": 1,
   "padding_side": "right",
-  "padding_value": 0.0,
+  "padding_value": 0,
+  "processor_class": "Wav2Vec2Processor",
   "return_attention_mask": true,
   "sampling_rate": 16000
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40f3119d7b387cde1143f032b8862b4730583a92e75d3234fa8907601400f994
+oid sha256:2cff9e72c664ff310fafc98b5ab498af5be5d4e2c23fdf2b40c8092c98f589b9
 size 6033