torgo_xlsr_finetune_M05_old – AI Model by macarious | AlphaNeural AI

You can deploy this model and start earning money today!

macarious/
torgo_xlsr_finetune_M05_old

like

0

automatic-speech-recognition

generated_from_trainer

endpoints_compatible

Views

No views yet

torgo_xlsr_finetune_M05

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.7932
Wer: 0.3577

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: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 8
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1000
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.5534	0.99	1000	3.4455	1.0
2.3481	1.98	2000	1.8194	0.8971
0.9664	2.97	3000	1.2685	0.6818
0.672	3.96	4000	1.3412	0.6112
0.5432	4.96	5000	1.4455	0.5275
0.4393	5.95	6000	1.3948	0.4761
0.3761	6.94	7000	1.8967	0.4785
0.3474	7.93	8000	1.5481	0.4545
0.309	8.92	9000	1.7275	0.4354
0.284	9.91	10000	1.9297	0.4438
0.2582	10.9	11000	1.4894	0.3971
0.2426	11.89	12000	1.6811	0.3840
0.2406	12.88	13000	1.7411	0.3935
0.2281	13.88	14000	1.7894	0.3732
0.1874	14.87	15000	1.7728	0.3864
0.1918	15.86	16000	2.0315	0.3768
0.1693	16.85	17000	1.7024	0.3672
0.1551	17.84	18000	1.7620	0.3684
0.1645	18.83	19000	1.7186	0.3696
0.1527	19.82	20000	1.7932	0.3577

Framework versions

Transformers 4.26.1
Pytorch 2.2.0
Datasets 2.16.1
Tokenizers 0.13.3