speecht5_finetuned_male_globo_add_token_3_facebook

This model was trained from scratch on the multilingual_librispeech dataset. It achieves the following results on the evaluation set:

• Loss: 0.3585

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: 1e-05
• train_batch_size: 4
• eval_batch_size: 2
• 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
• training_steps: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.4143	2.6	1000	0.3794
0.4056	5.2	2000	0.3701
0.3983	7.81	3000	0.3665
0.3988	10.41	4000	0.3641
0.3912	13.01	5000	0.3626
0.3901	15.61	6000	0.3604
0.3864	18.22	7000	0.3605
0.3895	20.82	8000	0.3591
0.3848	23.42	9000	0.3587
0.3868	26.02	10000	0.3585

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.1
• Datasets 2.18.0
• Tokenizers 0.15.2