Commit
•
49a61d3
1
Parent(s):
06523ed
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 |
+
{"ঁ": 1, "ং": 2, "ঃ": 3, "অ": 4, "আ": 5, "ই": 6, "ঈ": 7, "উ": 8, "এ": 9, "ঐ": 10, "ও": 11, "ঔ": 12, "ক": 13, "খ": 14, "গ": 15, "ঘ": 16, "ঙ": 17, "চ": 18, "ছ": 19, "জ": 20, "ঝ": 21, "ঞ": 22, "ট": 23, "ঠ": 24, "ড": 25, "ঢ": 26, "ণ": 27, "ত": 28, "থ": 29, "দ": 30, "ধ": 31, "ন": 32, "প": 33, "ফ": 34, "ব": 35, "ভ": 36, "ম": 37, "য": 38, "র": 39, "ল": 40, "শ": 41, "ষ": 42, "স": 43, "হ": 44, "়": 45, "া": 46, "ি": 47, "ী": 48, "ু": 49, "ূ": 50, "ৃ": 51, "ে": 52, "ৈ": 53, "ো": 54, "ৌ": 55, "্": 56, "ৎ": 57, "ড়": 58, "ঢ়": 59, "য়": 60, "ৰ": 61, "ৱ": 62, "|": 0, "[UNK]": 63, "[PAD]": 64}
|