URDU-ASR-LM

added_tokens.json

@@ -0,0 +1,3 @@
+{
+
+}

alphabet.json

@@ -0,0 +1 @@
+{"labels": ["2", "4", "\u062c", "\u062b", "\u0632", "\u0628", "\u0635", "\u06c1", "8", "\u0686", "\u0698", "5", "3", "\u062f", "\u062e", "7", "9", "\u06ba", "\u063a", "\u06cc", "\u06af", "\u0630", "\u0621", "\u062d", "\u0622", "\u0636", "1", "\u062a", "\u0644", "\u067e", "\u0645", "\u06a9", "0", "\u0633", "\u06be", "\u0679", "\u0688", "\u0637", "\u0646", "\u0638", "\u0639", "\u0627", "\u0642", " ", "6", "\u0648", "\u06d2", "\u0634", "\u0691", "\u0631", "\u0641", "\u2047", ""], "is_bpe": false}

language_model/5gram.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a632dcde356bfb9683c7fd869e80c964855241fc6b69cd30d15e52a4a1441b14
+size 42247717

language_model/attrs.json

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

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

@@ -1,6 +1 @@
-{
-  "bos_token": "<s>",
-  "eos_token": "</s>",
-  "pad_token": "[PAD]",
-  "unk_token": "[UNK]"
-}
+{"bos_token": null, "eos_token": null, "unk_token": "<UNK>", "pad_token": "<PAD>"}

tokenizer_config.json

@@ -2,9 +2,11 @@
   "bos_token": "<s>",
   "do_lower_case": false,
   "eos_token": "</s>",
+  "name_or_path": "urdumodel",
   "pad_token": "[PAD]",
-  "processor_class": "Wav2Vec2Processor",
+  "processor_class": "Wav2Vec2ProcessorWithLM",
   "replace_word_delimiter_char": " ",
+  "special_tokens_map_file": "urdumodel/special_tokens_map.json",
   "tokenizer_class": "Wav2Vec2CTCTokenizer",
   "unk_token": "[UNK]",
   "word_delimiter_token": "|"