Commit
•
6e4b53f
1
Parent(s):
690979e
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 |
+
{"ف": 0, "ٹ": 1, "ث": 2, "ط": 3, "ج": 4, "ء": 5, "ُ": 6, "د": 7, "ئ": 8, "ّ": 9, "ِ": 10, "ک": 11, "ہ": 12, "ڑ": 13, "پ": 14, "ق": 15, "س": 16, "ٔ": 17, "ض": 18, "ً": 19, "ص": 20, "ر": 21, "ب": 22, "خ": 23, "ت": 24, "چ": 25, "ژ": 26, "ذ": 27, "َ": 28, "گ": 29, "و": 30, "ے": 31, "ز": 32, "ں": 33, "ع": 34, "ش": 35, "غ": 36, "ڈ": 37, "آ": 38, "": 39, "ھ": 40, "ن": 41, "ل": 42, "ح": 43, "ؤ": 44, "ا": 45, "م": 46, "ی": 48, "ظ": 49, "|": 47, "[UNK]": 50, "[PAD]": 51}
|