update

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 46.77
+         value: 30.837004
 ---
 # Wav2Vec2-Large-XLSR-53-Japanese
 Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Japanese using the [Common Voice](https://huggingface.co/datasets/common_voice) and Japanese speech corpus of Saruwatari-lab, University of Tokyo [JSUT](https://sites.google.com/site/shinnosuketakamichi/publication/jsut).
@@ -107,7 +107,7 @@ def evaluate(batch):
 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**: 46.77
+**Test Result**: 30.837004
 ## Training
 The Common Voice `train`, `validation` datasets and Japanese speech corpus `basic5000` datasets were used for training.
-The script used for training can be found [here](Fine-Tune-Wav2Vec2-Large-XLSR-Japan.ipynb)
+The script used for training can be found [here](https://colab.research.google.com/drive/1ZTxoYzgOotUjcyoBf0m8gZj5Kcmu2yGU)