update model

README.md

@@ -24,10 +24,10 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 11.90
+         value: 9.36
        - name: Test CER
          type: cer
-         value: 2.94
+         value: 2.33
 ---
 
 # Wav2Vec2-Large-XLSR-53-Italian
@@ -49,7 +49,7 @@ from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 
 LANG_ID = "it"
 MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-italian"
-SAMPLES = 5
+SAMPLES = 10
 
 test_dataset = load_dataset("common_voice", LANG_ID, split=f"test[:{SAMPLES}]")
 
@@ -86,6 +86,11 @@ for i, predicted_sentence in enumerate(predicted_sentences):
 | "FIN DALL'INIZIO LA SEDE EPISCOPALE È STATA IMMEDIATAMENTE SOGGETTA ALLA SANTA SEDE." | FIN DALL'INIZIO LA SEDE EPISCOPALE È STATA IMMEDIATAMENTE SOGGETTA ALLA SANTA SEDE |
 | IL VUOTO ASSOLUTO? | IL VUOTO ASSOLUTO |
 | DOPO ALCUNI ANNI, EGLI DECISE DI TORNARE IN INDIA PER RACCOGLIERE ALTRI INSEGNAMENTI. | DOPO ALCUNI ANNI EGLI DECISE DI TORNARE IN INDIA PER RACCOGLIERE ALTRI INSEGNAMENTI |
+| SALVATION SUE | SALVATION SOO |
+| IN QUESTO MODO, DECIO OTTENNE IL POTERE IMPERIALE. | IN QUESTO MODO DECHO OTTENNE IL POTERE IMPERIALE |
+| SPARTA NOVARA ACQUISISCE IL TITOLO SPORTIVO PER GIOCARE IN PRIMA CATEGORIA. | PARCANOVARACFILISCE IL TITOLO SPORTIVO PER GIOCARE IN PRIMA CATEGORIA |
+| IN SEGUITO, KYGO E SHEAR HANNO PROPOSTO DI CONTINUARE A LAVORARE SULLA CANZONE. | IN SEGUITO KIGO E SHIAR HANNO PROPOSTO DI CONTINUARE A LAVORARE SULLA CANZONE |
+| ALAN CLARKE | ALAN CLARK |
 
 ## Evaluation
 
@@ -102,9 +107,11 @@ LANG_ID = "it"
 MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-italian"
 DEVICE = "cuda"
 
-CHARS_TO_IGNORE = [",", "?", "¿", ".", "!", "¡", ";", ":", '""', "%", '"', "�", "ʿ", "·", "჻", "~", "՞", 
+CHARS_TO_IGNORE = [",", "?", "¿", ".", "!", "¡", ";", "；", ":", '""', "%", '"', "�", "ʿ", "·", "჻", "~", "՞",
                    "؟", "،", "।", "॥", "«", "»", "„", "“", "”", "「", "」", "‘", "’", "《", "》", "(", ")", "[", "]",
-                   "=", "`", "_", "+", "<", ">", "…", "–", "°", "´", "ʾ", "‹", "›", "©", "®", "—", "→", "。"]
+                   "{", "}", "=", "`", "_", "+", "<", ">", "…", "–", "°", "´", "ʾ", "‹", "›", "©", "®", "—", "→", "。",
+                   "、", "﹂", "﹁", "‧", "～", "﹏", "，", "｛", "｝", "（", "）", "［", "］", "【", "】", "‥", "〽",
+                   "『", "』", "〝", "〟", "⟨", "⟩", "〜", "：", "！", "？", "♪", "؛", "/", "\\", "º", "−", "^", "ʻ", "ˆ"]
 
 test_dataset = load_dataset("common_voice", LANG_ID, split="test")
 
@@ -156,7 +163,7 @@ In the table below I report the Word Error Rate (WER) and the Character Error Ra
 
 | Model | WER | CER |
 | ------------- | ------------- | ------------- |
-| jonatasgrosman/wav2vec2-large-xlsr-53-italian | **11.90%** | **2.94%** |
+| jonatasgrosman/wav2vec2-large-xlsr-53-italian | **9.36%** | **2.33%** |
 | joorock12/wav2vec2-large-xlsr-italian | 12.60% | 3.18% |
 | gchhablani/wav2vec2-large-xlsr-it | 12.99% | 3.11% |
 | facebook/wav2vec2-large-xlsr-53-italian | 22.08% | 6.36% |

config.json

@@ -7,6 +7,8 @@
   ],
   "attention_dropout": 0.1,
   "bos_token_id": 1,
+  "codevector_dim": 768,
+  "contrastive_logits_temperature": 0.1,
   "conv_bias": true,
   "conv_dim": [
     512,
@@ -37,12 +39,14 @@
   ],
   "ctc_loss_reduction": "mean",
   "ctc_zero_infinity": true,
+  "diversity_loss_weight": 0.1,
   "do_stable_layer_norm": true,
   "eos_token_id": 2,
   "feat_extract_activation": "gelu",
   "feat_extract_dropout": 0.0,
   "feat_extract_norm": "layer",
   "feat_proj_dropout": 0.05,
+  "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
   "gradient_checkpointing": true,
   "hidden_act": "gelu",
@@ -66,11 +70,15 @@
   "mask_time_selection": "static",
   "model_type": "wav2vec2",
   "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
   "num_conv_pos_embedding_groups": 16,
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
   "num_hidden_layers": 24,
+  "num_negatives": 100,
   "pad_token_id": 0,
-  "transformers_version": "4.5.0.dev0",
-  "vocab_size": 75
+  "proj_codevector_dim": 768,
+  "transformers_version": "4.7.0.dev0",
+  "vocab_size": 44
 }

preprocessor_config.json

@@ -1,5 +1,6 @@
 {
   "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
   "feature_size": 1,
   "padding_side": "right",
   "padding_value": 0.0,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d9909cce880814662ccf4b45b2b3dcaf90801c7343acf2884e100ca50555b83
-size 1262241303
+oid sha256:14040a54d1e0caecd8b2271fdd2d96653be45702c7bc8765354140d864fb445a
+size 1262114199

vocab.json

@@ -1 +1 @@
-{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "Ð": 5, "Ó": 6, "Ł": 7, "Ô": 8, "O": 9, "Å": 10, "S": 11, "L": 12, "Ě": 13, "Ř": 14, "Ū": 15, "Ş": 16, "Ê": 17, "J": 18, "Q": 19, "Ō": 20, "U": 21, "Ž": 22, "A": 23, "Ć": 24, "B": 25, "Ï": 26, "Č": 27, "Ì": 28, "D": 29, "Y": 30, "Ğ": 31, "$": 32, "C": 33, "Á": 34, "P": 35, "Î": 36, "Ń": 37, "N": 38, "X": 39, "À": 40, "Ë": 41, "'": 42, "Ñ": 43, "Ī": 44, "È": 45, "Š": 46, "F": 47, "Ø": 48, "E": 49, "Ö": 50, "Ź": 51, "K": 52, "G": 53, "Ä": 54, "Ã": 55, "Μ": 56, "T": 57, "Ę": 58, "V": 59, "I": 60, "É": 61, "Z": 62, "R": 63, "-": 64, "W": 65, "Æ": 66, "M": 67, "Í": 68, "Œ": 69, "Ù": 70, "H": 71, "Ḥ": 72, "Ò": 73, "Ú": 74}
+{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "'": 5, "-": 6, "A": 7, "B": 8, "C": 9, "D": 10, "E": 11, "F": 12, "G": 13, "H": 14, "I": 15, "J": 16, "K": 17, "L": 18, "M": 19, "N": 20, "O": 21, "P": 22, "Q": 23, "R": 24, "S": 25, "T": 26, "U": 27, "V": 28, "W": 29, "X": 30, "Y": 31, "Z": 32, "À": 33, "Á": 34, "È": 35, "É": 36, "Ì": 37, "Í": 38, "Ò": 39, "Ó": 40, "Ù": 41, "Ú": 42, "Š": 43}