jimregan commited on
Commit
78d9d91
1 Parent(s): a8be5f0
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -18,7 +18,7 @@ model-index:
18
  dataset:
19
  name: Common Voice lv
20
  type: common_voice
21
- args: id
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 [Indonesian 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:
@@ -35,7 +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", "id", 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)
 
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)