update LM eval results

README.md

@@ -20,10 +20,16 @@ model-index:
       type: mozilla-foundation/common_voice_8_0
       args: ky
     metrics:
-       - name: Test WER
+       - name: Test WER (with LM)
+         type: wer
+         value: 19.01
+       - name: Test CER (with LM)
+         type: cer
+         value: 5.38
+       - name: Test WER (no LM)
          type: wer
          value: 31.28
-       - name: Test CER
+       - name: Test CER (no LM)
          type: cer
          value: 7.66
 ---
@@ -45,6 +51,8 @@ For a description of the model architecture, see [facebook/wav2vec2-xls-r-300m](
 
 The model vocabulary consists of the cyrillic alphabet with punctuation removed.
 
+The kenlm language model is built using the text of the train and invalidated corpus splits.
+
 ## Intended uses & limitations
 
 This model is expected to be of some utility for low-fidelity use cases such as:

mozilla-foundation_common_voice_8_0_ky_test_eval_results.txt

@@ -1,2 +1,2 @@
-WER: 0.27494497432134996
-CER: 0.0674079866403361
+WER: 0.19011371973587674
+CER: 0.05388927913480272