senga-nt-asr-inferred-force-aligned-speecht5-MAT-l1blend-0.7

This model is a fine-tuned version of microsoft/speecht5_tts on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0816

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: 3407
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 200
• num_epochs: 600.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.0825	58.8485	1000	0.0760
0.0736	117.6667	2000	0.0760
0.0614	176.4848	3000	0.0764
0.0578	235.3030	4000	0.0784
0.0552	294.1212	5000	0.0792
0.0553	352.9697	6000	0.0790
0.0515	411.7879	7000	0.0812
0.0482	470.6061	8000	0.0822
0.0507	529.4242	9000	0.0819
0.0493	588.2424	10000	0.0816

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu128
• Datasets 4.2.0
• Tokenizers 0.22.2