Vu Minh Chien
commited on
Commit
•
e249a0c
1
Parent(s):
b04a51b
update readme
Browse files
README.md
CHANGED
|
@@ -105,7 +105,7 @@ def evaluate(batch):
|
|
| 105 |
result = test_dataset.map(evaluate, batched=True, batch_size=8)
|
| 106 |
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
|
| 107 |
```
|
| 108 |
-
**Test Result**: 30.
|
| 109 |
## Training
|
| 110 |
The Common Voice `train`, `validation` datasets and Japanese speech corpus `basic5000` datasets were used for training.
|
| 111 |
The script used for training can be found [here](https://colab.research.google.com/drive/1ZTxoYzgOotUjcyoBf0m8gZj5Kcmu2yGU)
|
|
|
|
| 105 |
result = test_dataset.map(evaluate, batched=True, batch_size=8)
|
| 106 |
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
|
| 107 |
```
|
| 108 |
+
**Test Result**: 30.837%
|
| 109 |
## Training
|
| 110 |
The Common Voice `train`, `validation` datasets and Japanese speech corpus `basic5000` datasets were used for training.
|
| 111 |
The script used for training can be found [here](https://colab.research.google.com/drive/1ZTxoYzgOotUjcyoBf0m8gZj5Kcmu2yGU)
|