add model

config.json

@@ -0,0 +1,83 @@
+{
+  "activation_dropout": 0.0,
+  "apply_spec_augment": true,
+  "architectures": [
+    "UniSpeechForCTC"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 62,
+  "classifier_proj_size": 256,
+  "codevector_dim": 768,
+  "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,
+  "eos_token_id": 63,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.1,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_channel_length": 10,
+  "mask_channel_prob": 0.0,
+  "mask_feature_length": 10,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_space": 1,
+  "mask_time_prob": 0.075,
+  "model_type": "unispeech",
+  "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_ctc_classes": 80,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 24,
+  "num_negatives": 100,
+  "pad_token_id": 61,
+  "proj_codevector_dim": 768,
+  "replace_prob": 0.5,
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "torch_dtype": "float16",
+  "transformers_version": "4.13.0.dev0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 65
+}

preprocessor_config.json

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

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4543945a02416312ce57518f2566291dd3b99afc3a15b303c93907f98ca7ff17
+size 631197591

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>", "phonemizer_lang": "en-us", "phonemizer_backend": "espeak", "word_delimiter_token": "|", "do_lower_case": false, "tokenizer_class": "Wav2Vec2PhonemeCTCTokenizer"}

vocab.json

@@ -0,0 +1 @@
+{"e": 0, "tʃ": 1, "r": 2, "k": 3, "ɔ": 4, "d": 5, "ɪ": 6, "a": 7, "j": 8, "ʊ": 9, "t": 10, "i": 11, "l": 12, "p": 13, "ɡ": 14, "tː": 15, "n": 16, "o": 17, "ɛ": 18, "ɾ": 19, "m": 20, "u": 21, "s": 22, "iː": 23, "ss": 24, "b": 25, "v": 26, "dʒ": 27, "f": 28, "w": 29, "z": 30, "ʃ": 31, "ɲ": 32, "ʎ": 33, "eɪ": 34, "t̪": 35, "ts": 36, "pː": 37, "dʒː": 38, "dzː": 39, "ɔː": 40, "uː": 41, "kː": 42, "bː": 43, "aɪ": 44, "ŋ": 45, "tsː": 46, "dː": 47, "d̪": 48, "tʃː": 49, "aː": 50, "h": 51, "eː": 52, "ɛː": 53, "dz": 54, "oː": 55, "ɡː": 56, "ʒ": 57, "ɪː": 58, "oɪ": 59, "ʊː": 60, "<s>": 43, "<pad>": 42, "</s>": 63, "<unk>": 64}