RahelM/tigregna_tts1

TTS_tigregna

This model is a fine-tuned version of microsoft/speecht5_tts on the tigregna_20_hr dataset. It achieves the following results on the evaluation set:

• Loss: nan

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 20000

Training results

Training Loss	Epoch	Step	Validation Loss
0.0	11.33	1000	nan
0.0	22.66	2000	nan
0.0	33.99	3000	nan
0.0	45.33	4000	nan
0.0	56.66	5000	nan
0.0	67.99	6000	nan
0.0	79.32	7000	nan
0.0	90.65	8000	nan
0.0	101.98	9000	nan
0.0	113.31	10000	nan
0.0	124.65	11000	nan
0.0	135.98	12000	nan
0.0	147.31	13000	nan
0.0	158.64	14000	nan
0.0	169.97	15000	nan
0.0	181.3	16000	nan
0.0	192.63	17000	nan
0.0	203.97	18000	nan
0.0	215.3	19000	nan
0.0	226.63	20000	nan

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2