Commit
•
9be8971
1
Parent(s):
27b35f3
Update README.md
Browse files
README.md
CHANGED
@@ -51,4 +51,16 @@ for i, predicted_sentence in enumerate(predicted_sentences):
|
|
51 |
print("-" * 100)
|
52 |
print("Reference: ", test_dataset[i]["sentence"])
|
53 |
print("Prediction:", predicted_sentence)
|
54 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
51 |
print("-" * 100)
|
52 |
print("Reference: ", test_dataset[i]["sentence"])
|
53 |
print("Prediction:", predicted_sentence)
|
54 |
+
```
|
55 |
+
|
56 |
+
**Test Result**:
|
57 |
+
|
58 |
+
In the table below I report the Character Error Rate (CER) of the model tested on [TEDxJP-10K](https://github.com/laboroai/TEDxJP-10K) dataset.
|
59 |
+
|
60 |
+
| Model | CER |
|
61 |
+
| ------------- | ------------- |
|
62 |
+
| Ivydata/wav2vec2-large-xlsr-53-japanese | **27.87%** |
|
63 |
+
| jonatasgrosman/wav2vec2-large-xlsr-53-japanese | 34.18% |
|
64 |
+
| vumichien/wav2vec2-large-xlsr-japanese | 37.72% |
|
65 |
+
|
66 |
+
|