Upload lm-boosted decoder

alphabet.json

@@ -0,0 +1 @@
+{"labels": [" ", "-", "_", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "\u00a0", "\u00e5", "\u00e6", "\u00e9", "\u00f8", "\u2047", "", "<s>", "</s>"], "is_bpe": false}

language_model/5gram.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b41c24c63f2f0585bea83666369593f3b3e6d047f327a90f36ebca2c35ef0ff
+size 4243671427

language_model/attrs.json

@@ -0,0 +1 @@
+{"alpha": 0.5, "beta": 1.5, "unk_score_offset": -10.0, "score_boundary": true}

language_model/unigrams.txt

@@ -0,0 +1,22 @@
+a
+b
+c
+d
+e
+g
+h
+i
+k
+l
+m
+n
+o
+p
+r
+s
+t
+u
+v
+w
+x
+z

preprocessor_config.json

@@ -4,7 +4,7 @@
   "feature_size": 1,
   "padding_side": "right",
   "padding_value": 0.0,
-  "processor_class": "Wav2Vec2Processor",
+  "processor_class": "Wav2Vec2ProcessorWithLM",
   "return_attention_mask": true,
   "sampling_rate": 16000
 }

special_tokens_map.json

@@ -1,5 +1,19 @@
 {
   "additional_special_tokens": [
+    {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false
+    },
     {
       "content": "<s>",
       "lstrip": false,

tokenizer_config.json

@@ -3,9 +3,9 @@
   "do_lower_case": false,
   "eos_token": "</s>",
   "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "./",
+  "name_or_path": "scribe-project/wav2vec2-large-voxrex-300m-npsc_rundkast_nb_full",
   "pad_token": "[PAD]",
-  "processor_class": "Wav2Vec2Processor",
+  "processor_class": "Wav2Vec2ProcessorWithLM",
   "replace_word_delimiter_char": " ",
   "special_tokens_map_file": null,
   "tokenizer_class": "Wav2Vec2CTCTokenizer",