wav2vec2-large-xls-r-300m-AmhEnglish_last – AI Model by Mezge | AlphaNeural AI

You can deploy this model and start earning money today!

Mezge/
wav2vec2-large-xls-r-300m-AmhEnglish_last

like

0

generated_from_trainer

facebook/wav2vec2-xls-r-300m

Views

No views yet

wav2vec2-large-xls-r-300m-AmhEnglish_last

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.3828
Wer: 0.2116
Cer: 0.0560

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: 0.0003
train_batch_size: 4
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
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
num_epochs: 30
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
7.3981	1.88	400	3.4792	1.0	1.0
3.1249	3.76	800	2.2925	0.8205	0.5079
1.1402	5.63	1200	0.4348	0.3924	0.1125
0.3733	7.51	1600	0.3508	0.3265	0.0934
0.2274	9.39	2000	0.3322	0.2806	0.0786
0.1584	11.27	2400	0.3169	0.2625	0.0695
0.1174	13.15	2800	0.3233	0.2549	0.0673
0.0984	15.02	3200	0.3269	0.2443	0.0645
0.0832	16.9	3600	0.3427	0.2374	0.0630
0.0742	18.78	4000	0.3296	0.2370	0.0630
0.0612	20.66	4400	0.3588	0.2252	0.0599
0.0571	22.54	4800	0.3724	0.2233	0.0603
0.0494	24.41	5200	0.3787	0.2182	0.0583
0.0467	26.29	5600	0.3797	0.2158	0.0569
0.0387	28.17	6000	0.3828	0.2116	0.0560

Framework versions

Transformers 4.37.2
Pytorch 2.0.1+cu118
Datasets 3.6.0
Tokenizers 0.15.2