Update README.md
Browse files
README.md
CHANGED
@@ -25,7 +25,7 @@ model-index:
|
|
25 |
metrics:
|
26 |
- name: Test WER
|
27 |
type: wer
|
28 |
-
value:
|
29 |
---
|
30 |
# Wav2Vec2-Large-XLSR-53-Vietnamese
|
31 |
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Vietnamese using the [Common Voice](https://huggingface.co/datasets/common_voice), [FOSD](https://data.mendeley.com/datasets/k9sxg2twv4/4) and [VIVOS](https://ailab.hcmus.edu.vn/vivos).
|
@@ -312,7 +312,7 @@ def evaluate(batch):
|
|
312 |
result = test_dataset.map(evaluate, batched=True, batch_size=8)
|
313 |
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
|
314 |
```
|
315 |
-
**Test Result**: 49.
|
316 |
## Training
|
317 |
The Common Voice `train`, `validation` and FOSD datasets and VIVOS datasets were used for training as well.
|
318 |
The script used for training can be found [here](https://colab.research.google.com/drive/11pP4uVJj4SYZTzGjlCUtOHywlhYqs0cPx)
|
|
|
25 |
metrics:
|
26 |
- name: Test WER
|
27 |
type: wer
|
28 |
+
value: 49.59
|
29 |
---
|
30 |
# Wav2Vec2-Large-XLSR-53-Vietnamese
|
31 |
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Vietnamese using the [Common Voice](https://huggingface.co/datasets/common_voice), [FOSD](https://data.mendeley.com/datasets/k9sxg2twv4/4) and [VIVOS](https://ailab.hcmus.edu.vn/vivos).
|
|
|
312 |
result = test_dataset.map(evaluate, batched=True, batch_size=8)
|
313 |
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
|
314 |
```
|
315 |
+
**Test Result**: 49.59 %
|
316 |
## Training
|
317 |
The Common Voice `train`, `validation` and FOSD datasets and VIVOS datasets were used for training as well.
|
318 |
The script used for training can be found [here](https://colab.research.google.com/drive/11pP4uVJj4SYZTzGjlCUtOHywlhYqs0cPx)
|