speecht5_finetuned_pini_large
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: 0.5105
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: 15000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.5307 | 8.3333 | 100 | 0.4884 |
0.4795 | 16.6667 | 200 | 0.4629 |
0.457 | 25.0 | 300 | 0.4494 |
0.4467 | 33.3333 | 400 | 0.4486 |
0.4358 | 41.6667 | 500 | 0.4496 |
0.427 | 50.0 | 600 | 0.4461 |
0.4199 | 58.3333 | 700 | 0.4516 |
0.4112 | 66.6667 | 800 | 0.4445 |
0.4025 | 75.0 | 900 | 0.4463 |
0.4032 | 83.3333 | 1000 | 0.4538 |
0.3938 | 91.6667 | 1100 | 0.4527 |
0.3936 | 100.0 | 1200 | 0.4535 |
0.3861 | 108.3333 | 1300 | 0.4570 |
0.3819 | 116.6667 | 1400 | 0.4542 |
0.3832 | 125.0 | 1500 | 0.4564 |
0.3771 | 133.3333 | 1600 | 0.4575 |
0.3717 | 141.6667 | 1700 | 0.4547 |
0.3713 | 150.0 | 1800 | 0.4571 |
0.3678 | 158.3333 | 1900 | 0.4614 |
0.366 | 166.6667 | 2000 | 0.4581 |
0.363 | 175.0 | 2100 | 0.4606 |
0.3575 | 183.3333 | 2200 | 0.4572 |
0.3601 | 191.6667 | 2300 | 0.4635 |
0.3564 | 200.0 | 2400 | 0.4631 |
0.3563 | 208.3333 | 2500 | 0.4661 |
0.3496 | 216.6667 | 2600 | 0.4669 |
0.3491 | 225.0 | 2700 | 0.4645 |
0.3424 | 233.3333 | 2800 | 0.4695 |
0.3472 | 241.6667 | 2900 | 0.4709 |
0.3452 | 250.0 | 3000 | 0.4725 |
0.3427 | 258.3333 | 3100 | 0.4735 |
0.336 | 266.6667 | 3200 | 0.4697 |
0.3347 | 275.0 | 3300 | 0.4665 |
0.3345 | 283.3333 | 3400 | 0.4713 |
0.3313 | 291.6667 | 3500 | 0.4711 |
0.3328 | 300.0 | 3600 | 0.4739 |
0.331 | 308.3333 | 3700 | 0.4719 |
0.3324 | 316.6667 | 3800 | 0.4781 |
0.3273 | 325.0 | 3900 | 0.4727 |
0.3254 | 333.3333 | 4000 | 0.4789 |
0.3276 | 341.6667 | 4100 | 0.4720 |
0.3286 | 350.0 | 4200 | 0.4767 |
0.3224 | 358.3333 | 4300 | 0.4824 |
0.3255 | 366.6667 | 4400 | 0.4822 |
0.3296 | 375.0 | 4500 | 0.4787 |
0.3228 | 383.3333 | 4600 | 0.4806 |
0.3218 | 391.6667 | 4700 | 0.4765 |
0.321 | 400.0 | 4800 | 0.4820 |
0.3159 | 408.3333 | 4900 | 0.4815 |
0.3102 | 416.6667 | 5000 | 0.4859 |
0.3166 | 425.0 | 5100 | 0.4834 |
0.3133 | 433.3333 | 5200 | 0.4836 |
0.3125 | 441.6667 | 5300 | 0.4868 |
0.3096 | 450.0 | 5400 | 0.4869 |
0.3145 | 458.3333 | 5500 | 0.4846 |
0.3119 | 466.6667 | 5600 | 0.4871 |
0.3092 | 475.0 | 5700 | 0.4851 |
0.3087 | 483.3333 | 5800 | 0.4872 |
0.3027 | 491.6667 | 5900 | 0.4891 |
0.3075 | 500.0 | 6000 | 0.4911 |
0.3088 | 508.3333 | 6100 | 0.4874 |
0.3063 | 516.6667 | 6200 | 0.4891 |
0.3034 | 525.0 | 6300 | 0.4920 |
0.3021 | 533.3333 | 6400 | 0.4914 |
0.302 | 541.6667 | 6500 | 0.4893 |
0.3014 | 550.0 | 6600 | 0.4923 |
0.3004 | 558.3333 | 6700 | 0.4934 |
0.2995 | 566.6667 | 6800 | 0.4965 |
0.3014 | 575.0 | 6900 | 0.4918 |
0.3002 | 583.3333 | 7000 | 0.4926 |
0.3004 | 591.6667 | 7100 | 0.4970 |
0.2963 | 600.0 | 7200 | 0.4933 |
0.2974 | 608.3333 | 7300 | 0.4921 |
0.297 | 616.6667 | 7400 | 0.4958 |
0.2976 | 625.0 | 7500 | 0.4916 |
0.2959 | 633.3333 | 7600 | 0.4984 |
0.2971 | 641.6667 | 7700 | 0.4961 |
0.2955 | 650.0 | 7800 | 0.4938 |
0.2912 | 658.3333 | 7900 | 0.4969 |
0.2908 | 666.6667 | 8000 | 0.5002 |
0.2916 | 675.0 | 8100 | 0.4977 |
0.2905 | 683.3333 | 8200 | 0.4972 |
0.2926 | 691.6667 | 8300 | 0.4959 |
0.2901 | 700.0 | 8400 | 0.4983 |
0.2958 | 708.3333 | 8500 | 0.4977 |
0.2889 | 716.6667 | 8600 | 0.4998 |
0.2897 | 725.0 | 8700 | 0.4994 |
0.2886 | 733.3333 | 8800 | 0.5008 |
0.2877 | 741.6667 | 8900 | 0.4992 |
0.2864 | 750.0 | 9000 | 0.5032 |
0.2844 | 758.3333 | 9100 | 0.5001 |
0.2847 | 766.6667 | 9200 | 0.5003 |
0.2873 | 775.0 | 9300 | 0.4996 |
0.2841 | 783.3333 | 9400 | 0.5053 |
0.2861 | 791.6667 | 9500 | 0.5020 |
0.285 | 800.0 | 9600 | 0.4975 |
0.2849 | 808.3333 | 9700 | 0.5001 |
0.2895 | 816.6667 | 9800 | 0.4996 |
0.2826 | 825.0 | 9900 | 0.5018 |
0.2922 | 833.3333 | 10000 | 0.5039 |
0.2833 | 841.6667 | 10100 | 0.5043 |
0.2798 | 850.0 | 10200 | 0.5064 |
0.2852 | 858.3333 | 10300 | 0.5057 |
0.2809 | 866.6667 | 10400 | 0.5020 |
0.2833 | 875.0 | 10500 | 0.5042 |
0.2804 | 883.3333 | 10600 | 0.5011 |
0.2812 | 891.6667 | 10700 | 0.5038 |
0.2799 | 900.0 | 10800 | 0.5041 |
0.2784 | 908.3333 | 10900 | 0.5030 |
0.2779 | 916.6667 | 11000 | 0.5033 |
0.2811 | 925.0 | 11100 | 0.5072 |
0.2839 | 933.3333 | 11200 | 0.5047 |
0.2796 | 941.6667 | 11300 | 0.5046 |
0.2794 | 950.0 | 11400 | 0.5025 |
0.278 | 958.3333 | 11500 | 0.5063 |
0.278 | 966.6667 | 11600 | 0.5062 |
0.2765 | 975.0 | 11700 | 0.5075 |
0.2797 | 983.3333 | 11800 | 0.5061 |
0.2797 | 991.6667 | 11900 | 0.5102 |
0.276 | 1000.0 | 12000 | 0.5070 |
0.2759 | 1008.3333 | 12100 | 0.5063 |
0.2754 | 1016.6667 | 12200 | 0.5084 |
0.2783 | 1025.0 | 12300 | 0.5101 |
0.2784 | 1033.3333 | 12400 | 0.5078 |
0.28 | 1041.6667 | 12500 | 0.5089 |
0.2766 | 1050.0 | 12600 | 0.5076 |
0.277 | 1058.3333 | 12700 | 0.5092 |
0.2787 | 1066.6667 | 12800 | 0.5081 |
0.2727 | 1075.0 | 12900 | 0.5065 |
0.2736 | 1083.3333 | 13000 | 0.5081 |
0.2795 | 1091.6667 | 13100 | 0.5092 |
0.2767 | 1100.0 | 13200 | 0.5097 |
0.277 | 1108.3333 | 13300 | 0.5073 |
0.2786 | 1116.6667 | 13400 | 0.5083 |
0.2764 | 1125.0 | 13500 | 0.5066 |
0.275 | 1133.3333 | 13600 | 0.5089 |
0.2741 | 1141.6667 | 13700 | 0.5103 |
0.2718 | 1150.0 | 13800 | 0.5100 |
0.2775 | 1158.3333 | 13900 | 0.5097 |
0.2732 | 1166.6667 | 14000 | 0.5105 |
0.2729 | 1175.0 | 14100 | 0.5099 |
0.2746 | 1183.3333 | 14200 | 0.5102 |
0.279 | 1191.6667 | 14300 | 0.5108 |
0.2704 | 1200.0 | 14400 | 0.5101 |
0.2741 | 1208.3333 | 14500 | 0.5110 |
0.2731 | 1216.6667 | 14600 | 0.5124 |
0.2755 | 1225.0 | 14700 | 0.5105 |
0.2725 | 1233.3333 | 14800 | 0.5111 |
0.2773 | 1241.6667 | 14900 | 0.5109 |
0.2734 | 1250.0 | 15000 | 0.5105 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu124
- Datasets 2.21.0
- Tokenizers 0.19.1
- Downloads last month
- 2
Inference API (serverless) is not available, repository is disabled.