upload model

config.json

@@ -0,0 +1,81 @@
+{
+  "_name_or_path": "facebook/hubert-base-ls960",
+  "activation_dropout": 0.1,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 256,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": false,
+  "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": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.1,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.0,
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "mask_feature_length": 10,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_prob": 0.05,
+  "model_type": "wav2vec2",
+  "num_attention_heads": 12,
+  "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": 12,
+  "num_negatives": 100,
+  "pad_token_id": 38,
+  "proj_codevector_dim": 256,
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 39
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a8aafd2faa22e5a728aca3a1a78cedcc6b561f25731b738de839d7d361ecb16
+size 721728273

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
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c42247ac3fe8dd2d1b95cb51eafaf3baeb300d52d7966864af04823d1a896b56
+size 377691543

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2deea08dfafdcec82e0001c7eeb3ad52920cf67edeffc9370aa2d44a763dd3a4
+size 14567

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd07c1e7af0af38a2a25e0166936434897c50b2b9243d3b3f7693a359b580fec
+size 559

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66b0e7d13b8bc849e207ca234c1e1749af630b4667c008e039dbed6dedc440da
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

trainer_state.json

@@ -0,0 +1,133 @@
+{
+  "best_metric": 0.4794759750366211,
+  "best_model_checkpoint": "hubert-base-tamil-ipa/checkpoint-1500",
+  "epoch": 1.0706512042818912,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.91e-05,
+      "loss": 5.9666,
+      "step": 100
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 5.88e-05,
+      "loss": 2.691,
+      "step": 200
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 8.879999999999999e-05,
+      "loss": 2.5892,
+      "step": 300
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0001185,
+      "loss": 2.276,
+      "step": 400
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.0001482,
+      "loss": 0.9141,
+      "step": 500
+    },
+    {
+      "epoch": 0.36,
+      "eval_cer": 0.41183358071245185,
+      "eval_loss": 0.7568618655204773,
+      "eval_runtime": 1088.5554,
+      "eval_samples_per_second": 11.069,
+      "eval_steps_per_second": 1.384,
+      "step": 500
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0001779,
+      "loss": 0.5126,
+      "step": 600
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00020759999999999998,
+      "loss": 0.431,
+      "step": 700
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0002376,
+      "loss": 0.3866,
+      "step": 800
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.000267,
+      "loss": 0.3897,
+      "step": 900
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00029699999999999996,
+      "loss": 0.3504,
+      "step": 1000
+    },
+    {
+      "epoch": 0.71,
+      "eval_cer": 0.3094838133825417,
+      "eval_loss": 0.5331771373748779,
+      "eval_runtime": 1206.7225,
+      "eval_samples_per_second": 9.985,
+      "eval_steps_per_second": 1.249,
+      "step": 1000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00029900074019245003,
+      "loss": 0.3487,
+      "step": 1100
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0002979126572908956,
+      "loss": 0.3302,
+      "step": 1200
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.0002968245743893412,
+      "loss": 0.3246,
+      "step": 1300
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0002957142857142857,
+      "loss": 0.301,
+      "step": 1400
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00029460399703923015,
+      "loss": 0.2829,
+      "step": 1500
+    },
+    {
+      "epoch": 1.07,
+      "eval_cer": 0.2748139642169448,
+      "eval_loss": 0.4794759750366211,
+      "eval_runtime": 1193.809,
+      "eval_samples_per_second": 10.093,
+      "eval_steps_per_second": 1.262,
+      "step": 1500
+    }
+  ],
+  "max_steps": 28020,
+  "num_train_epochs": 20,
+  "total_flos": 2.852051261361667e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32c1264b4313409a32983632a06d3240d54f08adbbc04698a734ab29b0cfcf64
+size 2863

vocab.json

@@ -0,0 +1 @@
+{"h": 1, "aː": 2, "k": 3, "n̪": 4, "ɭ": 5, "p": 6, "ŋ": 7, "uː": 8, "்": 9, "ɾ": 10, "ɳ": 11, "i": 12, "d͡ʒ": 13, "t͡ʃ": 14, "m": 15, "n": 16, "r": 17, "oː": 18, "iː": 19, "t̪": 20, "eː": 21, "kʂ": 22, "ɻ": 23, "ʈ": 24, "o": 25, "ʂ": 26, "l": 27, "s": 28, "u": 29, "e": 30, "a": 31, "ʋ": 32, "aj": 33, "aʋ": 34, "j": 35, "ɲ": 36, "|": 0, "[UNK]": 37, "[PAD]": 38}