sanchit-gandhi
/

wav2vec2-2-gpt2-grid-search

Automatic Speech Recognition

speech-encoder-decoder

generated_from_trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

sanchit-gandhi HF staff commited on Feb 25, 2022

Commit

d581afb

•

1 Parent(s): fab98cf

update model card README.md

Files changed (1) hide show

README.md +5 -7

README.md CHANGED Viewed

@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 This model was trained from scratch on the librispeech_asr dataset.
 It achieves the following results on the evaluation set:
-- Loss: 4.3055
-- Wer: 1.9281
 ## Model description
@@ -39,8 +39,8 @@ The following hyperparameters were used during training:
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
@@ -51,9 +51,7 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Wer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 4.2675        | 0.28  | 500  | 4.6732          | 1.7896 |
-| 4.3772        | 0.56  | 1000 | 4.5290          | 1.9459 |
-| 4.1345        | 0.84  | 1500 | 4.3055          | 1.9281 |
 ### Framework versions

 This model was trained from scratch on the librispeech_asr dataset.
 It achieves the following results on the evaluation set:
+- Loss: 4.5379
+- Wer: 1.9297
 ## Model description
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
 | Training Loss | Epoch | Step | Validation Loss | Wer    |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
+| 4.1605        | 0.56  | 500  | 4.5379          | 1.9297 |
 ### Framework versions