Update README.md
Browse files
README.md
CHANGED
@@ -70,7 +70,7 @@ print("Reference:", test_dataset["sentence"][:2])
|
|
70 |
|
71 |
## Evaluation
|
72 |
|
73 |
-
The model can be evaluated as follows on the Luganda test data of Common Voice.
|
74 |
|
75 |
|
76 |
```python
|
@@ -87,7 +87,7 @@ processor = Wav2Vec2Processor.from_pretrained("lucio/wav2vec2-large-xlsr-luganda
|
|
87 |
model = Wav2Vec2ForCTC.from_pretrained("lucio/wav2vec2-large-xlsr-luganda")
|
88 |
model.to("cuda")
|
89 |
|
90 |
-
chars_to_ignore_regex = '[
|
91 |
resampler = torchaudio.transforms.Resample(48_000, 16_000)
|
92 |
|
93 |
# Preprocessing the datasets.
|
|
|
70 |
|
71 |
## Evaluation
|
72 |
|
73 |
+
The model can be evaluated as follows on the Luganda test data of Common Voice. (Available in Colab [here](https://colab.research.google.com/drive/1XxZ3mJOEXwIn-QH3C23jD_Qpom9aA1vH?usp=sharing).)
|
74 |
|
75 |
|
76 |
```python
|
|
|
87 |
model = Wav2Vec2ForCTC.from_pretrained("lucio/wav2vec2-large-xlsr-luganda")
|
88 |
model.to("cuda")
|
89 |
|
90 |
+
chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“]'
|
91 |
resampler = torchaudio.transforms.Resample(48_000, 16_000)
|
92 |
|
93 |
# Preprocessing the datasets.
|