WER

README.md

@@ -22,12 +22,12 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 15.91 
+         value: 15.80 
 ---
 
 # Wav2Vec2-Large-XLSR-53-German
 
-Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on German using 12% of the [Common Voice](https://huggingface.co/datasets/common_voice) dataset. 
+Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on German using the [Common Voice](https://huggingface.co/datasets/common_voice) dataset. 
 When using this model, make sure that your speech input is sampled at 16kHz.
 
 ## Usage
@@ -79,7 +79,7 @@ from datasets import load_dataset, load_metric
 from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
 import re
 
-test_dataset = load_dataset("common_voice", "de", split="test[:10%]")
+test_dataset = load_dataset("common_voice", "de", split="test")
 wer = load_metric("wer")
 
 processor = Wav2Vec2Processor.from_pretrained("marcel/wav2vec2-large-xlsr-53-german")
@@ -140,11 +140,10 @@ result = test_dataset.map(evaluate, batched=True, batch_size=8)
 print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
 ```
 
-**Test Result**: 15.91 % 
+**Test Result**: 15.80 % 
 
 
 ## Training
 
-The first 12% of the Common Voice `train`, `validation` datasets were used for training.
+The first 50% of the Common Voice `train`, and 12% of the `validation` datasets were used for training (30 epochs on first 12% and 3 epochs on the remainder).
 
-The script used for training can be found TODO