First model version (wiki-gl LM)
Browse files
README.md
CHANGED
|
@@ -61,3 +61,27 @@ model-index:
|
|
| 61 |
|
| 62 |
This is a copy of the model [diego-fustes/wav2vec2-large-xlsr-gl](https://huggingface.co/diego-fustes/wav2vec2-large-xlsr-gl) with an integrated language model.
|
| 63 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
|
| 62 |
This is a copy of the model [diego-fustes/wav2vec2-large-xlsr-gl](https://huggingface.co/diego-fustes/wav2vec2-large-xlsr-gl) with an integrated language model.
|
| 63 |
|
| 64 |
+
|
| 65 |
+
**Improvement**
|
| 66 |
+
This model has been compared with the baseline (diego-fustes/wav2vec2-large-xlsr-gl) on:
|
| 67 |
+
|
| 68 |
+
- The test subset of the Galician OpenSLR dataset (837 speech samples)
|
| 69 |
+
- The test subset of the Galician Common Voice 7.0 dataset (1716 speech samples)
|
| 70 |
+
|
| 71 |
+
The results are shown in the following tables:
|
| 72 |
+
|
| 73 |
+
OpenSLR77:
|
| 74 |
+
|
| 75 |
+
| Model | WER | CER |
|
| 76 |
+
| ------------- | ------------- | ------------- |
|
| 77 |
+
| diego-fustes/wav2vec2-large-xlsr-gl | **9.10%** | **3.94%** |
|
| 78 |
+
| cmagui/wav2vec2-large-xlsr-53-galician-with-lm | **6.86%** | **2.20%** |
|
| 79 |
+
|
| 80 |
+
Common_voice-gl:
|
| 81 |
+
|
| 82 |
+
| Model | WER | CER |
|
| 83 |
+
| ------------- | ------------- | ------------- |
|
| 84 |
+
| diego-fustes/wav2vec2-large-xlsr-gl | **22.12%** | **5.09%** |
|
| 85 |
+
| cmagui/wav2vec2-large-xlsr-53-galician-with-lm | **15.20%** | **3.87%** |
|
| 86 |
+
|
| 87 |
+
|