vc

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: 0.5712

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: 5e-06
• train_batch_size: 8
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• 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: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.9951	2.0851	50	0.9152
0.9256	4.1702	100	0.8480
0.9001	6.2553	150	0.8114
0.8567	8.3404	200	0.7857
0.8139	10.4255	250	0.7477
0.7437	12.5106	300	0.6724
0.6937	14.5957	350	0.6352
0.6686	16.6809	400	0.6194
0.6487	18.7660	450	0.6070
0.6411	20.8511	500	0.6009
0.643	22.9362	550	0.5970
0.6158	25.0	600	0.5893
0.632	27.0851	650	0.5871
0.6152	29.1702	700	0.5855
0.6066	31.2553	750	0.5833
0.615	33.3404	800	0.5817
0.6011	35.4255	850	0.5812
0.598	37.5106	900	0.5788
0.6023	39.5957	950	0.5764
0.6031	41.6809	1000	0.5775
0.5976	43.7660	1050	0.5764
0.597	45.8511	1100	0.5772
0.5923	47.9362	1150	0.5727
0.5793	50.0	1200	0.5746
0.5879	52.0851	1250	0.5757
0.5908	54.1702	1300	0.5727
0.5838	56.2553	1350	0.5745
0.5852	58.3404	1400	0.5709
0.5869	60.4255	1450	0.5753
0.585	62.5106	1500	0.5720
0.5875	64.5957	1550	0.5715
0.5807	66.6809	1600	0.5729
0.5886	68.7660	1650	0.5730
0.5831	70.8511	1700	0.5753
0.5812	72.9362	1750	0.5711
0.5736	75.0	1800	0.5768
0.5761	77.0851	1850	0.5735
0.5767	79.1702	1900	0.5759
0.5777	81.2553	1950	0.5720
0.5759	83.3404	2000	0.5712

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0