lm eval

.gitattributes

@@ -26,4 +26,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
+language_model/5gram.bin filter=lfs diff=lfs merge=lfs -text
 language_model/unigrams.txt filter=lfs diff=lfs merge=lfs -text

alphabet.json

@@ -0,0 +1 @@
+{"labels": [" ", "\u10d0", "\u10d1", "\u10d2", "\u10d3", "\u10d4", "\u10d5", "\u10d6", "\u10d7", "\u10d8", "\u10d9", "\u10da", "\u10db", "\u10dc", "\u10dd", "\u10de", "\u10df", "\u10e0", "\u10e1", "\u10e2", "\u10e3", "\u10e4", "\u10e5", "\u10e6", "\u10e7", "\u10e8", "\u10e9", "\u10ea", "\u10eb", "\u10ec", "\u10ed", "\u10ee", "\u10ef", "\u10f0", "\u2013", "\u2047", "", "<s>", "</s>"], "is_bpe": false}

eval.py

@@ -50,7 +50,7 @@ def log_results(result: Dataset, args: Dict[str, str]):
 def normalize_text(text: str) -> str:
     """This function normalizes the target text."""
 
-    chars_to_ignore_regex = re.compile("[^\sաբգդեզէըթժիլխծկհձղճմյնշոչպջռսվտրցւփքօֆև]")
+    chars_to_ignore_regex = re.compile("[^\sაბგდევზთიკლმნოპჟრსტუფქღყშჩცძწჭხჯჰ]")
     text = re.sub(chars_to_ignore_regex, "", text.lower())
     text = " ".join(text.split())
 

language_model/5gram.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24c77a100b046324791e2d207f3c806d0e068b3b4e2937777cceb0b308ebf584
+size 1663313932

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 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7853da7610a713cc14cb862a86ae602be9fea3d67352c9a20eeaad000c5dc21
+size 34592382

mozilla-foundation_common_voice_8_0_ka_test_eval_results.txt

@@ -0,0 +1,2 @@
+WER: 0.10356892932120364
+CER: 0.016409740438631165

preprocessor_config.json

@@ -3,7 +3,8 @@
   "feature_extractor_type": "Wav2Vec2FeatureExtractor",
   "feature_size": 1,
   "padding_side": "right",
-  "padding_value": 0,
+  "padding_value": 0.0,
+  "processor_class": "Wav2Vec2ProcessorWithLM",
   "return_attention_mask": true,
   "sampling_rate": 16000
 }

special_tokens_map.json

@@ -1 +1 @@
-{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

tokenizer_config.json

@@ -1 +1 @@
-{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
+{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "/workspace/output/ka/wav2vec2-xls-r-1b-ka", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2ProcessorWithLM"}