Commit ·
fde602a
1
Parent(s): c8addd9
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 |
+
{"ফ": 0, "ড": 1, "উ": 2, "৩": 3, "ধ": 4, "র": 5, "৬": 6, "চ": 7, "ঋ": 8, "ভ": 9, "ঝ": 10, "া": 11, "ঃ": 12, "ী": 13, "ে": 14, "়": 15, "ঢ়": 16, "স": 17, "২": 18, "গ": 19, "এ": 20, "ক": 21, "য়": 22, "ং": 23, "ট": 24, "জ": 25, "ই": 26, "ঠ": 27, "ৗ": 28, "ঞ": 29, "ঘ": 30, "।": 31, "৫": 32, "দ": 33, "০": 34, "ঢ": 35, "ও": 36, "আ": 37, "ঈ": 38, "ঊ": 39, "ষ": 40, "ণ": 41, "ম": 42, "৯": 43, " ": 44, "ছ": 45, "ৌ": 46, "ূ": 47, "ড়": 48, "ব": 49, "৭": 50, "ৰ": 51, "শ": 52, "ঙ": 53, "ত": 54, "ল": 55, "ন": 56, "অ": 57, "ৃ": 58, "ৎ": 59, "১": 60, "ি": 61, "ৈ": 62, "ঁ": 63, "ঔ": 64, "য": 65, "প": 66, "খ": 67, "ো": 68, "৪": 69, "ঐ": 70, "হ": 71, "ু": 72, "্": 73, "থ": 74, "৮": 75, "[UNK]": 76, "[PAD]": 77}
|