marma commited on
Commit
2d233c7
1 Parent(s): be08fa4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -2
README.md CHANGED
@@ -30,10 +30,11 @@ model-index:
30
  type: cer
31
  value: 3.946846
32
  ---
 
 
 
33
 
34
- **PLEASE NOTE [This](https://huggingface.co/KBLab/wav2vec2-large-voxrex-swedish) model performs better**
35
 
36
- # Wav2vec 2.0 large-voxpopuli-sv-swedish
37
  Additionally pretrained and finetuned version of Facebooks [VoxPopuli-sv large](https://huggingface.co/facebook/wav2vec2-large-sv-voxpopuli) model using Swedish radio broadcasts, NST and Common Voice data. Evalutation without a language model gives the following: WER for NST + Common Voice test set (2% of total sentences) is **3.95%**. WER for Common Voice test set is **10.99%** directly and **7.82%** with a 4-gram language model.
38
 
39
  When using this model, make sure that your speech input is sampled at 16kHz.
30
  type: cer
31
  value: 3.946846
32
  ---
33
+ # Wav2vec 2.0 large-voxpopuli-sv-swedish
34
+
35
+ **PLEASE NOTE that [this](https://huggingface.co/KBLab/wav2vec2-large-voxrex-swedish) model performs better and has a less restrictive license.**
36
 
 
37
 
 
38
  Additionally pretrained and finetuned version of Facebooks [VoxPopuli-sv large](https://huggingface.co/facebook/wav2vec2-large-sv-voxpopuli) model using Swedish radio broadcasts, NST and Common Voice data. Evalutation without a language model gives the following: WER for NST + Common Voice test set (2% of total sentences) is **3.95%**. WER for Common Voice test set is **10.99%** directly and **7.82%** with a 4-gram language model.
39
 
40
  When using this model, make sure that your speech input is sampled at 16kHz.