Commit ·
bd3a69e
1
Parent(s): 9637b79
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 |
+
{"ن": 0, "ب": 1, "ہ": 2, "ِ": 3, "ق": 4, "ے": 5, "ژ": 6, "ّ": 7, "ء": 8, "َ": 10, "ڑ": 11, "گ": 12, "ج": 13, "خ": 14, "ذ": 15, "ص": 16, "ث": 17, "ف": 18, "ٔ": 19, "ً": 20, "آ": 21, "ض": 22, "ئ": 23, "ا": 24, "س": 25, "ح": 26, "چ": 27, "ی": 28, "ں": 29, "ش": 30, "پ": 31, "و": 32, "ک": 33, "ز": 34, "ھ": 35, "ٹ": 36, "ڈ": 37, "د": 38, "ظ": 39, "ر": 40, "ع": 41, "ل": 42, "ط": 43, "غ": 44, "ُ": 45, "م": 46, "ت": 47, "ؤ": 48, "": 49, "|": 9, "[UNK]": 50, "[PAD]": 51}
|