VSydorskyi commited on
Commit
a95e066
1 Parent(s): 6d0ec1a

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +28 -0
README.md ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Model Card for Respeecher/ukrainian-data2vec
2
+
3
+ This model can be used as Feature Extractor model for Ukrainian language audio data
4
+
5
+ It can also be used as Backbone for downstream tasks, like ASR, Audio Classification, etc.
6
+
7
+ ### How to Get Started with the Model
8
+
9
+ ```python
10
+ from transformers import AutoProcessor, Data2VecAudioModel
11
+ import torch
12
+ from datasets import load_dataset, Audio
13
+
14
+ dataset = load_dataset("mozilla-foundation/common_voice_11_0", "uk", split="validation")
15
+ # Resample
16
+ dataset = dataset.cast_column("audio", Audio(sampling_rate=16_000))
17
+
18
+ processor = AutoProcessor.from_pretrained("Respeecher/ukrainian-data2vec")
19
+ model = Data2VecAudioModel.from_pretrained("Respeecher/ukrainian-data2vec")
20
+
21
+ # audio file is decoded on the fly
22
+ inputs = processor(dataset[0]["audio"]["array"], sampling_rate=sampling_rate, return_tensors="pt")
23
+ with torch.no_grad():
24
+ outputs = model(**inputs)
25
+
26
+ last_hidden_states = outputs.last_hidden_state
27
+ list(last_hidden_states.shape)
28
+ ```