vumichien
/

wav2vec2-large-xlsr-japanese

Automatic Speech Recognition

xlsr-fine-tuning-week

Inference Endpoints

Model card Files Files and versions Community

Vu Minh Chien commited on Apr 7, 2021

Commit

05014f2

•

1 Parent(s): dcbbaf5

update

Files changed (1) hide show

README.md +4 -2

README.md CHANGED Viewed

@@ -117,7 +117,9 @@ 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**: 30.837%
 ## 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](https://colab.research.google.com/drive/1ZTxoYzgOotUjcyoBf0m8gZj5Kcmu2yGU)

 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**
+**WER:** 30.84%
+**CER:** 17.85
 ## Training
 The Common Voice `train`, `validation` datasets and Japanese speech corpus `basic5000` datasets were used for training.