wav2vec2-1b-E30_freq_pause – AI Model by Gummybear05 | AlphaNeural AI

You can deploy this model and start earning money today!

Gummybear05/
wav2vec2-1b-E30_freq_pause

like

0

automatic-speech-recognition

generated_from_trainer

facebook/wav2vec2-xls-r-1b

endpoints_compatible

Views

No views yet

wav2vec2-1b-E30_freq_pause

This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.5438
Cer: 14.4032

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: 2
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
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: 50
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
8.581	0.2580	200	3.5252	80.8153
1.8592	0.5160	400	1.7795	40.3665
1.2512	0.7741	600	1.3170	32.6598
1.0721	1.0321	800	1.0213	24.8120
0.8561	1.2901	1000	0.9933	24.5242
0.7665	1.5481	1200	1.1026	26.9678
0.7108	1.8062	1400	0.8351	21.5519
0.6025	2.0642	1600	0.9202	23.6490
0.5026	2.3222	1800	0.8857	22.1511
0.4525	2.5802	2000	0.7021	18.6678
0.4254	2.8383	2200	0.7475	19.6722
0.3807	3.0963	2400	0.6803	17.8630
0.3106	3.3543	2600	0.6316	16.6177
0.2843	3.6123	2800	0.6282	16.8174
0.2704	3.8703	3000	0.6106	16.2829
0.2238	4.1284	3200	0.5700	14.9377
0.1923	4.3864	3400	0.5985	15.4429
0.1833	4.6444	3600	0.5486	14.6205
0.1779	4.9024	3800	0.5438	14.4032

Framework versions

Transformers 4.45.2
Pytorch 2.3.1.post100
Datasets 2.19.1
Tokenizers 0.20.1