Upload lm-boosted decoder
Browse files- alphabet.json +1 -0
- language_model/5gram_correct.arpa +0 -0
- language_model/attrs.json +1 -0
- language_model/unigrams.txt +91 -0
- preprocessor_config.json +2 -1
- tokenizer_config.json +2 -1
alphabet.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"labels": [" ", "#", "1", "2", "3", "4", "5", "a", "b", "c", "d", "e", "f", "g", "\u2047", "", "<s>", "</s>"], "is_bpe": false}
|
language_model/5gram_correct.arpa
ADDED
The diff for this file is too large to render.
See raw diff
|
|
language_model/attrs.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"alpha": 0.5, "beta": 1.5, "unk_score_offset": -10.0, "score_boundary": true}
|
language_model/unigrams.txt
ADDED
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
</s>
|
2 |
+
<s>
|
3 |
+
a#2
|
4 |
+
a#2.
|
5 |
+
a#3
|
6 |
+
a#3.
|
7 |
+
a#4
|
8 |
+
a#4.
|
9 |
+
a#5
|
10 |
+
a#5.
|
11 |
+
a1
|
12 |
+
a1.
|
13 |
+
a2
|
14 |
+
a2.
|
15 |
+
a3
|
16 |
+
a3.
|
17 |
+
a4
|
18 |
+
a4.
|
19 |
+
a5
|
20 |
+
b2
|
21 |
+
b2.
|
22 |
+
b3
|
23 |
+
b3.
|
24 |
+
b4
|
25 |
+
b4.
|
26 |
+
b5
|
27 |
+
c#3
|
28 |
+
c#3.
|
29 |
+
c#4
|
30 |
+
c#4.
|
31 |
+
c#5
|
32 |
+
c#5.
|
33 |
+
c3
|
34 |
+
c3.
|
35 |
+
c4
|
36 |
+
c4.
|
37 |
+
c5
|
38 |
+
c5.
|
39 |
+
d#3
|
40 |
+
d#3.
|
41 |
+
d#4
|
42 |
+
d#4.
|
43 |
+
d#5
|
44 |
+
d#5.
|
45 |
+
d2
|
46 |
+
d3
|
47 |
+
d3.
|
48 |
+
d4
|
49 |
+
d4.
|
50 |
+
d5
|
51 |
+
d5.
|
52 |
+
e2
|
53 |
+
e2.
|
54 |
+
e3
|
55 |
+
e3.
|
56 |
+
e4
|
57 |
+
e4.
|
58 |
+
e5
|
59 |
+
e5.
|
60 |
+
f#2
|
61 |
+
f#2.
|
62 |
+
f#3
|
63 |
+
f#3.
|
64 |
+
f#4
|
65 |
+
f#4.
|
66 |
+
f#5
|
67 |
+
f#5.
|
68 |
+
f1
|
69 |
+
f2
|
70 |
+
f3
|
71 |
+
f3.
|
72 |
+
f4
|
73 |
+
f4.
|
74 |
+
f5
|
75 |
+
f5.
|
76 |
+
g#2
|
77 |
+
g#2.
|
78 |
+
g#3
|
79 |
+
g#3.
|
80 |
+
g#4
|
81 |
+
g#4.
|
82 |
+
g#5
|
83 |
+
g#5.
|
84 |
+
g2
|
85 |
+
g2.
|
86 |
+
g3
|
87 |
+
g3.
|
88 |
+
g4
|
89 |
+
g4.
|
90 |
+
g5
|
91 |
+
g5.
|
preprocessor_config.json
CHANGED
@@ -3,7 +3,8 @@
|
|
3 |
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
4 |
"feature_size": 1,
|
5 |
"padding_side": "right",
|
6 |
-
"padding_value": 0,
|
|
|
7 |
"return_attention_mask": true,
|
8 |
"sampling_rate": 16000
|
9 |
}
|
|
|
3 |
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
4 |
"feature_size": 1,
|
5 |
"padding_side": "right",
|
6 |
+
"padding_value": 0.0,
|
7 |
+
"processor_class": "Wav2Vec2ProcessorWithLM",
|
8 |
"return_attention_mask": true,
|
9 |
"sampling_rate": 16000
|
10 |
}
|
tokenizer_config.json
CHANGED
@@ -2,12 +2,13 @@
|
|
2 |
"bos_token": "<s>",
|
3 |
"do_lower_case": false,
|
4 |
"eos_token": "</s>",
|
5 |
-
"name_or_path": "gary109/ai-light-dance_singing_ft_wav2vec2-large-lv60",
|
6 |
"pad_token": "[PAD]",
|
7 |
"processor_class": "Wav2Vec2ProcessorWithLM",
|
8 |
"replace_word_delimiter_char": " ",
|
9 |
"special_tokens_map_file": null,
|
10 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
|
|
11 |
"unk_token": "[UNK]",
|
12 |
"word_delimiter_token": "|"
|
13 |
}
|
|
|
2 |
"bos_token": "<s>",
|
3 |
"do_lower_case": false,
|
4 |
"eos_token": "</s>",
|
5 |
+
"name_or_path": "gary109/ai-light-dance_singing_ft_wav2vec2-large-lv60-v2",
|
6 |
"pad_token": "[PAD]",
|
7 |
"processor_class": "Wav2Vec2ProcessorWithLM",
|
8 |
"replace_word_delimiter_char": " ",
|
9 |
"special_tokens_map_file": null,
|
10 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
11 |
+
"trust_remote_code": false,
|
12 |
"unk_token": "[UNK]",
|
13 |
"word_delimiter_token": "|"
|
14 |
}
|