techiaith/microsoft_speecht5_finetuned_bu_tts_cy_en

speecht5_finetuned_bu_tts_cy_en

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: 0.4014

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: 32
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 256
• 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_steps: 500
• training_steps: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.4904	7.1942	1000	0.4423
0.4636	14.3885	2000	0.4227
0.4478	21.5827	3000	0.4148
0.4475	28.7770	4000	0.4094
0.4339	35.9712	5000	0.4061
0.4326	43.1655	6000	0.4047
0.4333	50.3597	7000	0.4033
0.4291	57.5540	8000	0.4031
0.4278	64.7482	9000	0.4016
0.4291	71.9424	10000	0.4014

Framework versions

• Transformers 4.50.2
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1