upd readme
Browse files
README.md
CHANGED
@@ -25,9 +25,9 @@ model-index:
|
|
25 |
value: 37.72
|
26 |
---
|
27 |
|
28 |
-
# Wav2Vec2-Large-XLSR-53-
|
29 |
|
30 |
-
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on
|
31 |
|
32 |
When using this model, make sure that your speech input is sampled at 16kHz.
|
33 |
|
@@ -42,7 +42,7 @@ import torchaudio
|
|
42 |
from datasets import load_dataset
|
43 |
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
|
44 |
|
45 |
-
test_dataset = load_dataset("common_voice", "
|
46 |
|
47 |
processor = Wav2Vec2Processor.from_pretrained("arampacha/wav2vec2-large-xlsr-ukrainian")
|
48 |
model = Wav2Vec2ForCTC.from_pretrained("arampacha/wav2vec2-large-xlsr-ukrainian")
|
@@ -82,11 +82,11 @@ from datasets import load_dataset, load_metric
|
|
82 |
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
|
83 |
import re
|
84 |
|
85 |
-
test_dataset = load_dataset("common_voice", "
|
86 |
|
87 |
wer = load_metric("wer")
|
88 |
-
processor = Wav2Vec2Processor.from_pretrained("arampacha/wav2vec2-large-xlsr-
|
89 |
-
model = Wav2Vec2ForCTC.from_pretrained("arampacha/wav2vec2-large-xlsr-
|
90 |
model.to("cuda")
|
91 |
|
92 |
chars_to_ignore = [",", "?", ".", "!", "-", ";", ":", '""', "%", "'", '"', "�", '«', '»', '—', '…', '(', ')', '*', '”', '“']
|
|
|
25 |
value: 37.72
|
26 |
---
|
27 |
|
28 |
+
# Wav2Vec2-Large-XLSR-53-Ukrainian
|
29 |
|
30 |
+
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Ukrainian using the [Common Voice](https://huggingface.co/datasets/common_voice) dataset.
|
31 |
|
32 |
When using this model, make sure that your speech input is sampled at 16kHz.
|
33 |
|
|
|
42 |
from datasets import load_dataset
|
43 |
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
|
44 |
|
45 |
+
test_dataset = load_dataset("common_voice", "uk", split="test[:2%]")
|
46 |
|
47 |
processor = Wav2Vec2Processor.from_pretrained("arampacha/wav2vec2-large-xlsr-ukrainian")
|
48 |
model = Wav2Vec2ForCTC.from_pretrained("arampacha/wav2vec2-large-xlsr-ukrainian")
|
|
|
82 |
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
|
83 |
import re
|
84 |
|
85 |
+
test_dataset = load_dataset("common_voice", "uk", split="test")
|
86 |
|
87 |
wer = load_metric("wer")
|
88 |
+
processor = Wav2Vec2Processor.from_pretrained("arampacha/wav2vec2-large-xlsr-ukrainian")
|
89 |
+
model = Wav2Vec2ForCTC.from_pretrained("arampacha/wav2vec2-large-xlsr-ukrainian")
|
90 |
model.to("cuda")
|
91 |
|
92 |
chars_to_ignore = [",", "?", ".", "!", "-", ";", ":", '""', "%", "'", '"', "�", '«', '»', '—', '…', '(', ')', '*', '”', '“']
|