update evaluation

README.md

@@ -22,16 +22,16 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 18.86
+         value: 10.38
        - name: Test CER
          type: cer
-         value: 6.16
+         value: 3.04
        - name: Test WER (+LM)
          type: wer
-         value: 13.53
+         value: 6.83
        - name: Test CER (+LM)
          type: cer
-         value: 5.24
+         value: 2.31
   - task: 
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
@@ -56,8 +56,12 @@ model-index:
 
 # XLS-R-1B-DUTCH
 
-This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - NL dataset.
+Fine-tuned [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on Dutch using the [Common Voice 8](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0).
+When using this model, make sure that your speech input is sampled at 16kHz.
 
+This model has been fine-tuned thanks to the GPU credits generously given by the [OVHcloud](https://www.ovhcloud.com/en/public-cloud/ai-training/) :)
+
+The script used for training can be found here: https://github.com/jonatasgrosman/wav2vec2-sprint
 
 ## Evaluation Commands
 

mozilla-foundation_common_voice_8_0_nl_test_eval_results.txt

@@ -1,2 +1,2 @@
-WER: 0.13539782050742627
-CER: 0.05246040201093512
+WER: 0.06834055623343291
+CER: 0.02318738499194393

mozilla-foundation_common_voice_8_0_nl_test_eval_results_greedy.txt

@@ -1,2 +1,2 @@
-WER: 0.18864391803761518
-CER: 0.061671611829651606
+WER: 0.10386249842218202
+CER: 0.030445769024203418