wav2vec_on_grid – AI Model by AntaraIIITD | AlphaNeural AI

You can deploy this model and start earning money today!

AntaraIIITD/
wav2vec_on_grid

like

0

automatic-speech-recognition

generated_from_trainer

endpoints_compatible

Views

No views yet

wav2vec_on_grid

This model is a fine-tuned version of facebook/wav2vec2-base-960h on the None dataset. It achieves the following results on the evaluation set:

Loss: -0.3810
Wer: 0.1345

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.1549	2.86	500	0.5346	0.5909
0.4979	5.71	1000	0.0507	0.4645
0.0943	8.57	1500	-0.1675	0.3417
-0.1114	11.43	2000	-0.3079	0.2426
-0.21	14.29	2500	-0.3226	0.2126
-0.2397	17.14	3000	-0.3238	0.2052
-0.2884	20.0	3500	-0.3635	0.1694
-0.307	22.86	4000	-0.3726	0.1540
-0.3311	25.71	4500	-0.3647	0.1520
-0.348	28.57	5000	-0.3889	0.1362
-0.3596	31.43	5500	-0.3726	0.1467
-0.366	34.29	6000	-0.3865	0.1364
-0.3758	37.14	6500	-0.3841	0.1398
-0.3876	40.0	7000	-0.3451	0.1485
-0.3983	42.86	7500	-0.3905	0.1350
-0.4015	45.71	8000	-0.3834	0.1362
-0.4067	48.57	8500	-0.3810	0.1345

Framework versions

Transformers 4.27.1
Pytorch 1.13.1+cu116
Datasets 2.10.1
Tokenizers 0.13.2