Upload tokenizer
Browse files- tokenizer_config.json +0 -1
- vocab.json +27 -27
tokenizer_config.json
CHANGED
|
@@ -3,7 +3,6 @@
|
|
| 3 |
"do_lower_case": false,
|
| 4 |
"eos_token": "</s>",
|
| 5 |
"pad_token": "[PAD]",
|
| 6 |
-
"processor_class": "Wav2Vec2Processor",
|
| 7 |
"replace_word_delimiter_char": " ",
|
| 8 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 9 |
"unk_token": "[UNK]",
|
|
|
|
| 3 |
"do_lower_case": false,
|
| 4 |
"eos_token": "</s>",
|
| 5 |
"pad_token": "[PAD]",
|
|
|
|
| 6 |
"replace_word_delimiter_char": " ",
|
| 7 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 8 |
"unk_token": "[UNK]",
|
vocab.json
CHANGED
|
@@ -1,32 +1,32 @@
|
|
| 1 |
{
|
| 2 |
-
"'":
|
| 3 |
"[PAD]": 29,
|
| 4 |
"[UNK]": 28,
|
| 5 |
-
"a":
|
| 6 |
-
"b":
|
| 7 |
-
"c":
|
| 8 |
-
"d":
|
| 9 |
-
"e":
|
| 10 |
-
"f":
|
| 11 |
-
"g":
|
| 12 |
-
"h":
|
| 13 |
"i": 15,
|
| 14 |
-
"j":
|
| 15 |
-
"k":
|
| 16 |
-
"l":
|
| 17 |
-
"m":
|
| 18 |
-
"n":
|
| 19 |
-
"o":
|
| 20 |
-
"p":
|
| 21 |
-
"q":
|
| 22 |
-
"r":
|
| 23 |
-
"s":
|
| 24 |
-
"t":
|
| 25 |
-
"u":
|
| 26 |
-
"v":
|
| 27 |
-
"w":
|
| 28 |
-
"x":
|
| 29 |
-
"y":
|
| 30 |
-
"z":
|
| 31 |
-
"|":
|
| 32 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"'": 1,
|
| 3 |
"[PAD]": 29,
|
| 4 |
"[UNK]": 28,
|
| 5 |
+
"a": 3,
|
| 6 |
+
"b": 7,
|
| 7 |
+
"c": 19,
|
| 8 |
+
"d": 2,
|
| 9 |
+
"e": 6,
|
| 10 |
+
"f": 13,
|
| 11 |
+
"g": 24,
|
| 12 |
+
"h": 16,
|
| 13 |
"i": 15,
|
| 14 |
+
"j": 26,
|
| 15 |
+
"k": 27,
|
| 16 |
+
"l": 12,
|
| 17 |
+
"m": 14,
|
| 18 |
+
"n": 10,
|
| 19 |
+
"o": 21,
|
| 20 |
+
"p": 18,
|
| 21 |
+
"q": 11,
|
| 22 |
+
"r": 5,
|
| 23 |
+
"s": 0,
|
| 24 |
+
"t": 8,
|
| 25 |
+
"u": 17,
|
| 26 |
+
"v": 22,
|
| 27 |
+
"w": 23,
|
| 28 |
+
"x": 25,
|
| 29 |
+
"y": 4,
|
| 30 |
+
"z": 9,
|
| 31 |
+
"|": 20
|
| 32 |
}
|