gwella'r WER

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 27.21
+         value: 25.31
 ---
 
 # Wav2Vec2-Large-XLSR-Welsh
@@ -122,7 +122,7 @@ result = test_dataset.map(evaluate, batched=True, batch_size=8)
 print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
 ```
 
-**Test Result**: 27.21% 
+**Test Result**: 25.31% 
 
 
 # Training 

config.json

@@ -1,11 +1,11 @@
 {
   "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
-  "activation_dropout": 0.0,
+  "activation_dropout": 0.055,
   "apply_spec_augment": true,
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.1,
+  "attention_dropout": 0.094,
   "bos_token_id": 1,
   "conv_bias": true,
   "conv_dim": [
@@ -42,16 +42,16 @@
   "feat_extract_activation": "gelu",
   "feat_extract_dropout": 0.0,
   "feat_extract_norm": "layer",
-  "feat_proj_dropout": 0.0,
+  "feat_proj_dropout": 0.04,
   "final_dropout": 0.0,
   "gradient_checkpointing": true,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.1,
+  "hidden_dropout": 0.047,
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.1,
+  "layerdrop": 0.041,
   "mask_channel_length": 10,
   "mask_channel_min_space": 1,
   "mask_channel_other": 0.0,
@@ -62,7 +62,7 @@
   "mask_time_length": 10,
   "mask_time_min_space": 1,
   "mask_time_other": 0.0,
-  "mask_time_prob": 0.05,
+  "mask_time_prob": 0.082,
   "mask_time_selection": "static",
   "model_type": "wav2vec2",
   "num_attention_heads": 16,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:acb3d9a52a1f16dd818c7191f1d292bb027e8d277756d823a0071794721add4a
-size 1262130515
+oid sha256:0b3d1d4c5339a82fac76d5a361323ba7c8e8faa278c8f37e1e4db9929ecd4705
+size 1262129841

vocab.json

@@ -1 +1 @@
-{"a": 0, "ñ": 1, "ä": 2, "u": 3, "x": 4, "n": 5, "¬": 6, "ï": 7, "ô": 8, "k": 9, "á": 10, "h": 11, "v": 12, "ÿ": 13, "r": 14, "m": 15, "é": 16, "q": 17, "à": 18, "w": 19, "j": 20, "g": 21, "z": 22, "d": 23, "ŷ": 24, "c": 25, "s": 26, "l": 27, "ö": 28, "î": 29, "e": 30, "b": 31, "‘": 32, "o": 33, "â": 34, "t": 35, "ò": 36, "i": 37, "y": 38, "ŵ": 39, "p": 40, "ê": 41, "ë": 42, "f": 43, "'": 44, "û": 46, "|": 45, "[UNK]": 47, "[PAD]": 48}
+{"g": 1, "‘": 2, "e": 3, "r": 4, "b": 5, "ê": 6, "â": 7, "ŷ": 8, "'": 9, "v": 10, "ö": 11, "ï": 12, "h": 13, "z": 14, "m": 15, "á": 16, "ä": 17, "ñ": 18, "o": 19, "n": 20, "j": 21, "ò": 22, "ë": 23, "î": 24, "k": 25, "q": 26, "x": 27, "¬": 28, "a": 29, "s": 30, "i": 31, "ÿ": 32, "u": 33, "ŵ": 34, "c": 35, "é": 36, "w": 37, "p": 38, "y": 39, "d": 40, "l": 41, "à": 42, "û": 43, "t": 44, "f": 45, "ô": 46, "|": 0, "[UNK]": 47, "[PAD]": 48}