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