Training in progress, step 200

config.json

@@ -0,0 +1,107 @@
+{
+  "activation_dropout": 0.05,
+  "adapter_attn_dim": 16,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 1024,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.05,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.0,
+  "hidden_size": 1280,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "model_type": "wav2vec2",
+  "num_adapter_layers": 3,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 48,
+  "num_negatives": 100,
+  "output_hidden_size": 1280,
+  "pad_token_id": 57,
+  "proj_codevector_dim": 1024,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "transformers_version": "4.57.2",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 60,
+  "xvector_output_dim": 512
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e12e34f0183d5de0b7eb249a292379659f89786631f22c154bc0c07b46726b6d
+size 3859039520

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04defc10cea1e72f6299dd544b9114920da9841c9bf809b41c8ddb90301e30e9
+size 6097

vocab.json

@@ -1,62 +1 @@
-{
-  "ttj": {
-    "&": 9,
-    "'": 21,
-    "-": 0,
-    "1": 35,
-    "A": 43,
-    "B": 29,
-    "C": 47,
-    "D": 26,
-    "E": 31,
-    "F": 55,
-    "G": 22,
-    "H": 36,
-    "I": 15,
-    "K": 27,
-    "L": 38,
-    "M": 40,
-    "N": 3,
-    "O": 34,
-    "P": 2,
-    "Q": 49,
-    "R": 17,
-    "S": 50,
-    "T": 53,
-    "U": 13,
-    "W": 39,
-    "Y": 19,
-    "[PAD]": 57,
-    "[UNK]": 56,
-    "a": 45,
-    "b": 4,
-    "bb": 14,
-    "c": 16,
-    "d": 33,
-    "e": 8,
-    "f": 5,
-    "g": 7,
-    "h": 10,
-    "i": 25,
-    "j": 6,
-    "k": 24,
-    "l": 11,
-    "m": 28,
-    "n": 37,
-    "ny": 30,
-    "o": 44,
-    "p": 54,
-    "r": 18,
-    "s": 52,
-    "t": 32,
-    "u": 23,
-    "v": 51,
-    "w": 12,
-    "y": 20,
-    "z": 46,
-    "|": 48,
-    "ö": 41,
-    "ü": 42,
-    "ę": 1
-  }
-}
+{"ttj": {"-": 0, "\u0119": 1, "P": 2, "N": 3, "b": 4, "f": 5, "j": 6, "g": 7, "e": 8, "&": 9, "h": 10, "l": 11, "w": 12, "U": 13, "bb": 14, "I": 15, "c": 16, "R": 17, "r": 18, "Y": 19, "y": 20, "'": 21, "G": 22, "u": 23, "k": 24, "i": 25, "D": 26, "K": 27, "m": 28, "B": 29, "ny": 30, "E": 31, "t": 32, "d": 33, "O": 34, "1": 35, "H": 36, "n": 37, "L": 38, "W": 39, "M": 40, "\u00f6": 41, "\u00fc": 42, "A": 43, "o": 44, "a": 45, "z": 46, "C": 47, "Q": 49, "S": 50, "v": 51, "s": 52, "T": 53, "p": 54, "F": 55, "|": 48, "[UNK]": 56, "[PAD]": 57}}