Update README.md

README.md

@@ -49,14 +49,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 # 
 
-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.
+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. This model is also available with a language model which improves these results. This model can be found at https://huggingface.co/RuudVelo/wav2vec2-large-xls-r-1b-nl-lm. The Common Voice 8 Dutch test Wer is 9.73 of that model. 
 It achieves the following results on the evaluation set:
 - Loss: 0.1479
 - Wer: 0.1156
 
 ## Model description
 
-More information needed
+Model fine-tuned using the wav2vec-als-r-1b model architecture
 
 ## Intended uses & limitations
 
@@ -64,24 +64,13 @@ More information needed
 
 ## Training and evaluation data
 
-More information needed
+Model has been trained on Common Voice 8 Dutch
 
 ## Training procedure
 
 ### Training hyperparameters
 
-The following hyperparameters were used during training:
-- learning_rate: 5e-05
-- train_batch_size: 8
-- eval_batch_size: 8
-- seed: 42
-- gradient_accumulation_steps: 5
-- total_train_batch_size: 40
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 500
-- num_epochs: 50.0
-- mixed_precision_training: Native AMP
+Model parameters can be found under Files and versions in the run.sh file.
 
 ### Training results
 
@@ -189,5 +178,5 @@ The following hyperparameters were used during training:
 
 - Transformers 4.17.0.dev0
 - Pytorch 1.10.2+cu102
-- Datasets 1.18.2.dev0
+- Datasets 1.18.3
 - Tokenizers 0.11.0