patrickvonplaten
commited on
Commit
·
18eefaa
1
Parent(s):
63a73ec
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 |
+
{"r": 0, "v": 1, "ğ": 2, "w": 3, "n": 4, "̇": 5, "ü": 6, "'": 7, "j": 8, "x": 9, "ë": 10, "â": 11, "p": 12, "ı": 13, "e": 14, "s": 15, "o": 16, "y": 17, "t": 18, "c": 19, "ş": 20, "ö": 21, "i": 22, "u": 23, "f": 24, "ç": 25, "k": 26, "m": 27, "q": 28, "z": 29, "l": 30, "î": 31, "h": 32, "g": 33, "b": 35, "a": 36, "d": 37, "|": 34, "[UNK]": 38, "[PAD]": 39}
|