Millad

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

• Loss: 3.2265
• Wer: 0.5465
• Cer: 0.3162

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: 8
• 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: 4000
• num_epochs: 750
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
3.2911	33.9	2000	2.2097	0.9963	0.6047
1.3419	67.8	4000	1.9042	0.7007	0.3565
0.6542	101.69	6000	1.7195	0.5985	0.3194
0.373	135.59	8000	2.2219	0.6078	0.3241
0.2805	169.49	10000	2.3114	0.6320	0.3304
0.2014	203.39	12000	2.6898	0.6338	0.3597
0.1611	237.29	14000	2.7808	0.6041	0.3379
0.1265	271.19	16000	2.8304	0.5632	0.3289
0.1082	305.08	18000	2.8373	0.5874	0.3344
0.103	338.98	20000	2.8580	0.5743	0.3292
0.0854	372.88	22000	2.5413	0.5539	0.3186
0.0675	406.78	24000	2.5523	0.5502	0.3229
0.0531	440.68	26000	2.9369	0.5483	0.3142
0.0504	474.58	28000	3.1416	0.5595	0.3225
0.0388	508.47	30000	2.5655	0.5390	0.3111
0.0396	542.37	32000	3.1923	0.5558	0.3178
0.0274	576.27	34000	2.9235	0.5520	0.3257
0.0361	610.17	36000	3.3828	0.5762	0.3312
0.02	644.07	38000	3.3822	0.5874	0.3466
0.0176	677.97	40000	3.1191	0.5539	0.3209
0.0181	711.86	42000	3.2022	0.5576	0.3237
0.0124	745.76	44000	3.2265	0.5465	0.3162

Framework versions

• Transformers 4.17.0
• Pytorch 1.12.0+cu113
• Datasets 1.18.3
• Tokenizers 0.12.1