first_release

README.md

@@ -1,6 +1,6 @@
 ---
 license: apache-2.0
-base_model: facebook/wav2vec2-xls-r-300m
+base_model: facebook/wav2vec2-large-xlsr-53
 tags:
 - generated_from_trainer
 datasets:
@@ -15,15 +15,15 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Wav2Vec2-Urdu-300M
 
-This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice_13_0 dataset.
+This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the common_voice_13_0 dataset.
 It achieves the following results on the evaluation set:
 - eval_loss: inf
-- eval_wer: 0.6044
-- eval_runtime: 235.4174
-- eval_samples_per_second: 14.035
-- eval_steps_per_second: 1.754
-- epoch: 5.17
-- step: 4800
+- eval_wer: 0.3928
+- eval_runtime: 234.5067
+- eval_samples_per_second: 14.089
+- eval_steps_per_second: 1.761
+- epoch: 13.56
+- step: 8400
 
 ## Model description
 
@@ -43,11 +43,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0003
-- train_batch_size: 4
+- train_batch_size: 6
 - eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 8
+- total_train_batch_size: 12
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44145407c64cfff4be1adad5db1f255a453a773bc541c736784a486214edc6b5
+oid sha256:2f1f359573045800386d975fe5f4899628b42d2d623fdb74f62045f693af05f3
 size 1261950980

runs/Nov06_16-13-33_638c50298a65/events.out.tfevents.1699287443.638c50298a65.6792.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e1facc720d58324b06df127381c6478c4635ba7d50517675d493161b0f2ec852
-size 10937
+oid sha256:54238a411b685e0865c64b5696e0bde2e69adc92bebd9816c5c34fd16a02dc78
+size 11412