add generated json

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "processor_class": "Wav2Vec2Processor",
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token": "<s>",
+  "do_lower_case": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "processor_class": "Wav2Vec2Processor",
+  "replace_word_delimiter_char": " ",
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "unk_token": "<unk>",
+  "word_delimiter_token": "|"
+}

vocab.json

@@ -0,0 +1,89 @@
+{
+  "</s>": 2,
+  "<pad>": 0,
+  "<s>": 1,
+  "<unk>": 3,
+  "a": 4,
+  "ai": 5,
+  "au": 6,
+  "bʲ": 7,
+  "bˠ": 8,
+  "c": 9,
+  "d": 10,
+  "dʲ": 11,
+  "d̠ʲ": 12,
+  "d̪ˠ": 13,
+  "eː": 14,
+  "fʲ": 15,
+  "fˠ": 16,
+  "hʲ": 17,
+  "hˠ": 18,
+  "ia": 19,
+  "iə": 20,
+  "iː": 21,
+  "j": 22,
+  "k": 23,
+  "l": 24,
+  "lʲ": 25,
+  "l̠̊ʲ": 26,
+  "l̪̊ˠ": 27,
+  "l̠ʲ": 28,
+  "l̥": 29,
+  "l̥ʲ": 30,
+  "l̪ˠ": 31,
+  "mʲ": 32,
+  "mˠ": 33,
+  "m̥ʲ": 34,
+  "m̥ˠ": 35,
+  "n": 36,
+  "nʲ": 37,
+  "n̠̊ʲ": 38,
+  "n̪̊ˠ": 39,
+  "n̠ʲ": 40,
+  "n̥ʲ": 41,
+  "n̪ˠ": 42,
+  "oː": 43,
+  "pʲ": 44,
+  "pˠ": 45,
+  "sˠ": 46,
+  "t": 47,
+  "tʲ": 48,
+  "t̠ʲ": 49,
+  "t̪ˠ": 50,
+  "ua": 51,
+  "uə": 52,
+  "uː": 53,
+  "vʲ": 54,
+  "vˠ": 55,
+  "w": 56,
+  "x": 57,
+  "z": 58,
+  "|": 59,
+  "æː": 60,
+  "ç": 61,
+  "ŋ": 62,
+  "ŋ̊": 63,
+  "ɑː": 64,
+  "ɔu": 65,
+  "ɔː": 66,
+  "ɕ": 67,
+  "ə": 68,
+  "ɛ": 69,
+  "ɟ": 70,
+  "ɡ": 71,
+  "ɣ": 72,
+  "ɤ": 73,
+  "ɪ": 74,
+  "ɲ": 75,
+  "ɾʲ": 76,
+  "ɾˠ": 77,
+  "ɾ̥ʲ": 78,
+  "ɾ̥ˠ": 79,
+  "ʃ": 80,
+  "ʉː": 81,
+  "ʊ": 82,
+  "ʌ": 83,
+  "ʝ": 84,
+  "ʤ": 85,
+  "ʧ": 86
+}