zainulhakim commited on
Commit
7e49100
1 Parent(s): e01f660

Upload tokenizer

Browse files
Files changed (3) hide show
  1. README.md +3 -3
  2. tokenizer_config.json +0 -1
  3. vocab.json +23 -23
README.md CHANGED
@@ -1,10 +1,10 @@
1
  ---
2
- license: apache-2.0
3
  base_model: facebook/wav2vec2-base
4
- tags:
5
- - generated_from_trainer
6
  metrics:
7
  - wer
 
 
8
  model-index:
9
  - name: 240615-wav2vec2-ASR-English
10
  results: []
 
1
  ---
 
2
  base_model: facebook/wav2vec2-base
3
+ license: apache-2.0
 
4
  metrics:
5
  - wer
6
+ tags:
7
+ - generated_from_trainer
8
  model-index:
9
  - name: 240615-wav2vec2-ASR-English
10
  results: []
tokenizer_config.json CHANGED
@@ -39,7 +39,6 @@
39
  "eos_token": "</s>",
40
  "model_max_length": 1000000000000000019884624838656,
41
  "pad_token": "[PAD]",
42
- "processor_class": "Wav2Vec2Processor",
43
  "replace_word_delimiter_char": " ",
44
  "target_lang": null,
45
  "tokenizer_class": "Wav2Vec2CTCTokenizer",
 
39
  "eos_token": "</s>",
40
  "model_max_length": 1000000000000000019884624838656,
41
  "pad_token": "[PAD]",
 
42
  "replace_word_delimiter_char": " ",
43
  "target_lang": null,
44
  "tokenizer_class": "Wav2Vec2CTCTokenizer",
vocab.json CHANGED
@@ -1,28 +1,28 @@
1
  {
2
  "[PAD]": 25,
3
  "[UNK]": 24,
4
- "a": 4,
5
- "b": 7,
6
- "c": 23,
7
- "d": 21,
8
- "e": 18,
9
- "f": 11,
10
- "g": 5,
11
- "h": 1,
12
  "i": 6,
13
- "k": 3,
14
- "l": 17,
15
- "m": 22,
16
- "n": 16,
17
- "o": 9,
18
- "p": 8,
19
- "r": 19,
20
- "s": 12,
21
- "t": 14,
22
- "u": 15,
23
- "v": 10,
24
- "w": 20,
25
- "x": 13,
26
- "y": 2,
27
- "|": 0
28
  }
 
1
  {
2
  "[PAD]": 25,
3
  "[UNK]": 24,
4
+ "a": 22,
5
+ "b": 2,
6
+ "c": 15,
7
+ "d": 13,
8
+ "e": 11,
9
+ "f": 7,
10
+ "g": 21,
11
+ "h": 18,
12
  "i": 6,
13
+ "k": 20,
14
+ "l": 10,
15
+ "m": 8,
16
+ "n": 0,
17
+ "o": 5,
18
+ "p": 4,
19
+ "r": 3,
20
+ "s": 9,
21
+ "t": 17,
22
+ "u": 1,
23
+ "v": 23,
24
+ "w": 12,
25
+ "x": 19,
26
+ "y": 14,
27
+ "|": 16
28
  }