speecht5_female_british_english_speaker_2

This model is a fine-tuned version of microsoft/speecht5_tts on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4968

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: 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: 100
• training_steps: 1100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.4995	7.9208	100	0.4349
0.421	15.8416	200	0.4270
0.3934	23.7624	300	0.4447
0.3765	31.6832	400	0.4632
0.3665	39.6040	500	0.4657
0.3595	47.5248	600	0.4665
0.3455	55.4455	700	0.4821
0.3405	63.3663	800	0.4851
0.3349	71.2871	900	0.4916
0.3289	79.2079	1000	0.4928
0.329	87.1287	1100	0.4968

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1