add tokenizer
Browse files- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- vocab.json +1 -0
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "replace_word_delimiter_char": " ", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"8": 0, "â": 1, "g": 2, "2": 3, "1": 4, "l": 5, "a": 6, ")": 7, "f": 8, "k": 9, "n": 10, "b": 11, "u": 12, "®": 13, "r": 14, "_": 15, "s": 16, "q": 17, "i": 18, "d": 19, "'": 20, "0": 21, "w": 22, "*": 23, "j": 24, "9": 25, "#": 26, "p": 27, "x": 28, "y": 29, "&": 30, "3": 31, "e": 32, "z": 33, "6": 34, "o": 36, "h": 37, "@": 38, "4": 39, "c": 40, "(": 41, "/": 42, "7": 43, "m": 44, "%": 45, "5": 46, "+": 47, "v": 48, "t": 49, "|": 35, "[UNK]": 50, "[PAD]": 51}
|