You can deploy this model and start earning money today!

nik1509/
telugu_wav2vec_hiddendropout_ablation_hd_03

Name: telugu_wav2vec_hiddendropout_ablation_hd_03
Availability: InStock

transformers

safetensors

wav2vec2

automatic-speech-recognition

generated_from_trainer

nik1509/telugu_wav2vecmedium_basemodel

finetune

apache-2.0

endpoints_compatible

Views

No views yet

telugu_wav2vec_hiddendropout_ablation_hd_03

This model is a fine-tuned version of nik1509/telugu_wav2vecmedium_basemodel on the None dataset. It achieves the following results on the evaluation set:

Loss: 583.5934
Wer: 0.3946
Cer: 0.1692

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: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 15
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
336.5259	0.6076	700	454.9762	0.4135	0.1711
275.6354	1.2153	1400	472.1844	0.4061	0.1686
288.2869	1.8229	2100	468.2711	0.4076	0.1694
285.0092	2.4306	2800	474.0866	0.4075	0.1693
286.3706	3.0382	3500	487.9208	0.4038	0.1671
227.2902	3.6458	4200	485.6910	0.3986	0.1680
253.1653	4.2535	4900	494.0355	0.4030	0.1680
285.7851	4.8611	5600	498.2196	0.3995	0.1673
252.8042	5.4688	6300	500.8943	0.4012	0.1679
260.7867	6.0764	7000	514.0763	0.3910	0.1661
267.5607	6.6840	7700	511.0818	0.3958	0.1671
278.1148	7.2917	8400	520.3400	0.3951	0.1671
266.0618	7.8993	9100	520.4923	0.3943	0.1676
222.7104	8.5069	9800	529.5754	0.3947	0.1680
238.997	9.1146	10500	539.8677	0.3955	0.1683
279.4554	9.7222	11200	539.7169	0.3977	0.1685
246.4711	10.3299	11900	549.1530	0.3963	0.1680
227.0257	10.9375	12600	546.2502	0.3940	0.1683
257.3041	11.5451	13300	555.3053	0.3958	0.1683
207.4065	12.1528	14000	571.1086	0.3968	0.1688
262.2076	12.7604	14700	565.4132	0.3951	0.1691
220.2876	13.3681	15400	577.4109	0.3919	0.1684
251.8584	13.9757	16100	575.3079	0.3964	0.1687
202.1257	14.5833	16800	583.5934	0.3946	0.1692

Framework versions

Transformers 4.52.4
Pytorch 2.8.0.dev20250609+cu118
Datasets 3.6.0
Tokenizers 0.21.2

nik1509/telugu_wav2vec_hiddendropout_ablation_hd_03

telugu_wav2vec_hiddendropout_ablation_hd_03

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

nik1509/
telugu_wav2vec_hiddendropout_ablation_hd_03