Update README with CV18 eval results

README.md

@@ -13,8 +13,8 @@ model-index:
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: Common Voice 17.0
-      type: mozilla-foundation/common_voice_17_0
+      name: Common Voice 18.0
+      type: mozilla-foundation/common_voice_18_0
       config: eu
       split: test
       args:
@@ -22,7 +22,7 @@ model-index:
     metrics:
       - name: Test WER
         type: wer
-        value: 4.47
+        value: 4.84
 base_model:
 - xezpeleta/whisper-large-v3-eu
 ---
@@ -37,7 +37,7 @@ base_model:
 - **Dataset:** `asierhv/composite_corpus_eu_v2.1` (Mozilla Common Voice 18.0 + Basque Parliament + OpenSLR)
 - **Conversion Format:** CTranslate2 (optimized for inference)
 - **Compatibility:** Designed for use with [faster-whisper](https://github.com/SYSTRAN/faster-whisper)
-- **WER:** 4.47% on Mozilla Common Voice 17.0
+- **WER:** 4.84% on Mozilla Common Voice 18.0
 
 ## Usage with faster-whisper
 
@@ -64,7 +64,7 @@ for segment in segments:
 
 ## Evaluation
 
-The model achieves **4.47% Word Error Rate (WER)** on the Basque `test` split of **Mozilla Common Voice 17.0**.
+The model achieves **4.84% Word Error Rate (WER)** on the Basque `test` split of **Mozilla Common Voice 18.0**.
 
 
 # Conversion details