Commit
·
6a540cd
1
Parent(s):
5b8f935
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 |
+
{"ِ": 0, "ف": 1, "ژ": 2, "ح": 3, "گ": 4, "و": 5, "خ": 6, "ے": 8, "ت": 9, "َ": 10, "ھ": 11, "ڈ": 12, "م": 13, "ر": 14, "ُ": 15, "ط": 16, "ض": 17, "ق": 18, "س": 19, "ذ": 20, "ص": 21, "د": 22, "ا": 23, "ظ": 24, "ً": 25, "پ": 26, "غ": 27, "ج": 28, "ہ": 29, "ّ": 30, "": 31, "ش": 32, "ث": 33, "ب": 34, "ٔ": 35, "ک": 36, "ل": 37, "ں": 38, "چ": 39, "ڑ": 40, "ز": 41, "ئ": 42, "ٹ": 43, "ؤ": 44, "آ": 45, "ء": 46, "ی": 47, "ن": 48, "ع": 49, "|": 7, "[UNK]": 50, "[PAD]": 51}
|