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": "|", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"ʃ": 1, "l": 2, "aj": 3, "s": 4, "n": 5, "θ": 6, "ej": 7, "ow": 8, "m": 9, "i": 10, "t": 11, "ʒ": 12, "ð": 13, "ɹ": 14, "ɛ": 15, "t͡ʃ": 16, "ɔ": 17, "h": 18, "ɪ": 19, "oj": 20, "aw": 21, "ɹ̩": 22, "ɑ": 23, "ʊ": 24, "j": 25, "d": 26, "v": 27, "ŋ": 28, "z": 29, "d͡ʒ": 30, "f": 31, "ʌ": 32, "k": 33, "w": 34, "u": 35, "ɡ": 36, "ə": 37, "b": 38, "p": 39, "æ": 40, "|": 0, "[UNK]": 41, "[PAD]": 42}
|