habiba
/

egy-slang-model

Automatic Speech Recognition

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

habiba commited on Jan 12, 2022

Commit

0f1097b

•

1 Parent(s): ff713a6

update model card README.md

Files changed (1) hide show

README.md +15 -12

README.md CHANGED Viewed

@@ -14,8 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.9538
-- Wer: 0.9996
 ## Model description
@@ -43,22 +43,25 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
-- num_epochs: 30
 - mixed_precision_training: Native AMP
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Wer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 3.7978        | 3.28  | 400  | 2.9727          | 1.0    |
-| 2.9541        | 6.56  | 800  | 2.9667          | 1.0    |
-| 2.9529        | 9.83  | 1200 | 2.9739          | 0.9999 |
-| 2.9228        | 13.11 | 1600 | 2.9690          | 0.9999 |
-| 2.8879        | 16.39 | 2000 | 3.0275          | 1.0    |
-| 2.8703        | 19.67 | 2400 | 3.0226          | 0.9999 |
-| 2.8552        | 22.95 | 2800 | 2.9886          | 0.9999 |
-| 2.8397        | 26.23 | 3200 | 2.9590          | 0.9999 |
-| 2.8026        | 29.51 | 3600 | 2.9538          | 0.9996 |
 ### Framework versions

 This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 2.9273
+- Wer: 1.0000
 ## Model description
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
+- num_epochs: 20
 - mixed_precision_training: Native AMP
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Wer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
+| No log        | 1.64  | 200  | 2.9735          | 1.0    |
+| 3.8098        | 3.28  | 400  | 2.9765          | 1.0    |
+| 3.8098        | 4.91  | 600  | 2.9662          | 1.0    |
+| 2.9531        | 6.56  | 800  | 2.9708          | 1.0    |
+| 2.9531        | 8.2   | 1000 | 2.9673          | 1.0    |
+| 2.9259        | 9.83  | 1200 | 2.9989          | 1.0    |
+| 2.9259        | 11.47 | 1400 | 2.9889          | 1.0    |
+| 2.9023        | 13.11 | 1600 | 2.9739          | 1.0    |
+| 2.9023        | 14.75 | 1800 | 3.0040          | 1.0000 |
+| 2.8832        | 16.39 | 2000 | 3.0170          | 1.0    |
+| 2.8832        | 18.03 | 2200 | 2.9963          | 0.9999 |
+| 2.8691        | 19.67 | 2400 | 2.9273          | 1.0000 |
 ### Framework versions