Upload processor
Browse files- preprocessor_config.json +10 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +14 -0
- vocab.json +86 -0
preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"processor_class": "Wav2Vec2Processor",
|
| 8 |
+
"return_attention_mask": false,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"clean_up_tokenization_spaces": true,
|
| 4 |
+
"do_lower_case": false,
|
| 5 |
+
"do_normalize": false,
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 8 |
+
"pad_token": "<pad>",
|
| 9 |
+
"processor_class": "Wav2Vec2Processor",
|
| 10 |
+
"return_attention_mask": false,
|
| 11 |
+
"tokenizer_class": "Wav2Vec2Tokenizer",
|
| 12 |
+
"unk_token": "<unk>",
|
| 13 |
+
"word_delimiter_token": "|"
|
| 14 |
+
}
|
vocab.json
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"0000101": 0,
|
| 3 |
+
"0000110": 1,
|
| 4 |
+
"0000111": 2,
|
| 5 |
+
"0001001": 3,
|
| 6 |
+
"0001010": 4,
|
| 7 |
+
"0001011": 5,
|
| 8 |
+
"0001101": 6,
|
| 9 |
+
"0001110": 7,
|
| 10 |
+
"0001111": 8,
|
| 11 |
+
"0010001": 9,
|
| 12 |
+
"0010010": 10,
|
| 13 |
+
"0010011": 11,
|
| 14 |
+
"0010101": 12,
|
| 15 |
+
"0010110": 13,
|
| 16 |
+
"0010111": 14,
|
| 17 |
+
"0011001": 15,
|
| 18 |
+
"0011010": 16,
|
| 19 |
+
"0011011": 17,
|
| 20 |
+
"0011101": 18,
|
| 21 |
+
"0011110": 19,
|
| 22 |
+
"0011111": 20,
|
| 23 |
+
"0100001": 21,
|
| 24 |
+
"0100010": 22,
|
| 25 |
+
"0100011": 23,
|
| 26 |
+
"0100101": 24,
|
| 27 |
+
"0100110": 25,
|
| 28 |
+
"0100111": 26,
|
| 29 |
+
"0101001": 27,
|
| 30 |
+
"0101010": 28,
|
| 31 |
+
"0101011": 29,
|
| 32 |
+
"0101101": 30,
|
| 33 |
+
"0101110": 31,
|
| 34 |
+
"0101111": 32,
|
| 35 |
+
"0110001": 33,
|
| 36 |
+
"0110010": 34,
|
| 37 |
+
"0110011": 35,
|
| 38 |
+
"0110101": 36,
|
| 39 |
+
"0110110": 37,
|
| 40 |
+
"0110111": 38,
|
| 41 |
+
"0111001": 39,
|
| 42 |
+
"0111010": 40,
|
| 43 |
+
"0111011": 41,
|
| 44 |
+
"0111101": 42,
|
| 45 |
+
"0111110": 43,
|
| 46 |
+
"0111111": 44,
|
| 47 |
+
"1000001": 45,
|
| 48 |
+
"1000010": 46,
|
| 49 |
+
"1000011": 47,
|
| 50 |
+
"1000101": 48,
|
| 51 |
+
"1000110": 49,
|
| 52 |
+
"1000111": 50,
|
| 53 |
+
"1001001": 51,
|
| 54 |
+
"1001010": 52,
|
| 55 |
+
"1001011": 53,
|
| 56 |
+
"1001101": 54,
|
| 57 |
+
"1001110": 55,
|
| 58 |
+
"1001111": 56,
|
| 59 |
+
"1010001": 57,
|
| 60 |
+
"1010010": 58,
|
| 61 |
+
"1010011": 59,
|
| 62 |
+
"1010101": 60,
|
| 63 |
+
"1010110": 61,
|
| 64 |
+
"1010111": 62,
|
| 65 |
+
"1011001": 63,
|
| 66 |
+
"1011010": 64,
|
| 67 |
+
"1011011": 65,
|
| 68 |
+
"1011101": 66,
|
| 69 |
+
"1011110": 67,
|
| 70 |
+
"1011111": 68,
|
| 71 |
+
"1100001": 69,
|
| 72 |
+
"1100010": 70,
|
| 73 |
+
"1100011": 71,
|
| 74 |
+
"1100101": 72,
|
| 75 |
+
"1100110": 73,
|
| 76 |
+
"1100111": 74,
|
| 77 |
+
"1101001": 75,
|
| 78 |
+
"1101010": 76,
|
| 79 |
+
"1101011": 77,
|
| 80 |
+
"1101101": 78,
|
| 81 |
+
"1101110": 79,
|
| 82 |
+
"1101111": 80,
|
| 83 |
+
"1110001": 81,
|
| 84 |
+
"1110010": 82,
|
| 85 |
+
"1110011": 83
|
| 86 |
+
}
|