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 |
+
{"و": 0, "ٹ": 1, "ج": 2, "د": 3, "س": 4, "ذ": 5, "ً": 6, "ء": 7, "چ": 8, "ق": 9, "": 10, "ل": 11, "ع": 12, "ص": 13, "ا": 14, "ڑ": 15, "ز": 16, "ُ": 17, "خ": 18, "ح": 19, "ظ": 20, "ھ": 21, "ڈ": 22, "ے": 24, "ت": 25, "ں": 26, "ط": 27, "ر": 28, "ش": 29, "م": 30, "آ": 31, "گ": 32, "ک": 33, "پ": 34, "ّ": 35, "ث": 36, "ئ": 37, "ٔ": 38, "ف": 39, "ب": 40, "ض": 41, "ن": 42, "ِ": 43, "غ": 44, "ی": 45, "ہ": 46, "ژ": 47, "ؤ": 48, "َ": 49, "|": 23, "[UNK]": 50, "[PAD]": 51}
|