fix more
Browse files
README.md
CHANGED
@@ -18,7 +18,7 @@ model-index:
|
|
18 |
dataset:
|
19 |
name: Common Voice lv
|
20 |
type: common_voice
|
21 |
-
args:
|
22 |
metrics:
|
23 |
- name: Test WER
|
24 |
type: wer
|
@@ -26,7 +26,7 @@ model-index:
|
|
26 |
---
|
27 |
# Wav2Vec2-Large-XLSR-Latvian
|
28 |
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53)
|
29 |
-
on the [
|
30 |
When using this model, make sure that your speech input is sampled at 16kHz.
|
31 |
## Usage
|
32 |
The model can be used directly (without a language model) as follows:
|
@@ -35,7 +35,7 @@ import torch
|
|
35 |
import torchaudio
|
36 |
from datasets import load_dataset
|
37 |
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
|
38 |
-
test_dataset = load_dataset("common_voice", "
|
39 |
processor = Wav2Vec2Processor.from_pretrained("jimregan/wav2vec2-large-xlsr-latvian-cv")
|
40 |
model = Wav2Vec2ForCTC.from_pretrained("jimregan/wav2vec2-large-xlsr-latvian-cv")
|
41 |
resampler = torchaudio.transforms.Resample(48_000, 16_000)
|
|
|
18 |
dataset:
|
19 |
name: Common Voice lv
|
20 |
type: common_voice
|
21 |
+
args: lv
|
22 |
metrics:
|
23 |
- name: Test WER
|
24 |
type: wer
|
|
|
26 |
---
|
27 |
# Wav2Vec2-Large-XLSR-Latvian
|
28 |
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53)
|
29 |
+
on the [Latvian Common Voice dataset](https://huggingface.co/datasets/common_voice).
|
30 |
When using this model, make sure that your speech input is sampled at 16kHz.
|
31 |
## Usage
|
32 |
The model can be used directly (without a language model) as follows:
|
|
|
35 |
import torchaudio
|
36 |
from datasets import load_dataset
|
37 |
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
|
38 |
+
test_dataset = load_dataset("common_voice", "lv", split="test[:2%]")
|
39 |
processor = Wav2Vec2Processor.from_pretrained("jimregan/wav2vec2-large-xlsr-latvian-cv")
|
40 |
model = Wav2Vec2ForCTC.from_pretrained("jimregan/wav2vec2-large-xlsr-latvian-cv")
|
41 |
resampler = torchaudio.transforms.Resample(48_000, 16_000)
|