--- license: cc-by-nc-4.0 base_model: facebook/mms-1b-all tags: - generated_from_trainer metrics: - wer model-index: - name: mms-MGB3 results: [] --- # mms-MGB3 This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8109 - Wer: 56.2371 ## 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: 14 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: constant_with_warmup - lr_scheduler_warmup_steps: 50 - num_epochs: 25 ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:-----:|:---------------:|:-------:| | 9.7535 | 0.13 | 250 | 8.6735 | 1.0023 | | 3.2385 | 0.27 | 500 | 3.3341 | 1.0003 | | 2.3512 | 0.4 | 750 | 2.0937 | 0.9027 | | 1.4967 | 0.53 | 1000 | 1.3694 | 0.7637 | | 1.3214 | 0.67 | 1250 | 1.2237 | 0.7347 | | 1.2072 | 0.8 | 1500 | 1.1672 | 0.7176 | | 1.1913 | 0.93 | 1750 | 1.1334 | 0.7108 | | 1.1127 | 1.07 | 2000 | 1.1102 | 0.7044 | | 1.1454 | 1.2 | 2250 | 1.0919 | 0.6996 | | 1.1128 | 1.33 | 2500 | 1.0763 | 0.6955 | | 1.086 | 1.47 | 2750 | 1.0629 | 0.6916 | | 1.1285 | 1.6 | 3000 | 1.0503 | 0.6888 | | 1.081 | 1.73 | 3250 | 1.0406 | 0.6886 | | 1.0449 | 1.86 | 3500 | 1.0320 | 0.6857 | | 1.0625 | 2.0 | 3750 | 1.0231 | 0.6849 | | 1.0892 | 2.13 | 4000 | 1.0157 | 0.6824 | | 1.0566 | 2.26 | 4250 | 1.0097 | 0.6795 | | 1.0972 | 2.4 | 4500 | 1.0036 | 0.6747 | | 1.0617 | 2.53 | 4750 | 0.9957 | 0.6744 | | 1.0441 | 2.66 | 5000 | 0.9881 | 0.6756 | | 1.0589 | 2.8 | 5250 | 0.9807 | 0.6718 | | 1.0005 | 2.93 | 5500 | 0.9758 | 0.6713 | | 1.0447 | 3.06 | 5750 | 0.9701 | 0.6694 | | 0.9722 | 3.2 | 6000 | 0.9667 | 0.6664 | | 0.9873 | 3.33 | 6250 | 0.9595 | 0.6675 | | 0.9857 | 3.46 | 6500 | 0.9551 | 0.6633 | | 0.9625 | 3.6 | 6750 | 0.9519 | 0.6633 | | 0.9748 | 3.73 | 7000 | 0.9464 | 0.6607 | | 0.9626 | 3.86 | 7250 | 0.9427 | 0.6617 | | 1.0242 | 4.0 | 7500 | 0.9382 | 0.6591 | | 0.9704 | 4.13 | 7750 | 0.9358 | 65.5829 | | 0.9154 | 4.26 | 8000 | 0.9332 | 65.4060 | | 0.9491 | 4.4 | 8250 | 0.9290 | 65.8311 | | 0.9419 | 4.53 | 8500 | 0.9230 | 65.2067 | | 0.931 | 4.66 | 8750 | 0.9197 | 65.2331 | | 0.9614 | 4.79 | 9000 | 0.9175 | 64.7236 | | 0.9214 | 4.93 | 9250 | 0.9145 | 64.8991 | | 0.9268 | 5.06 | 9500 | 0.9110 | 64.6153 | | 0.9124 | 5.19 | 9750 | 0.9082 | 64.4226 | | 0.9357 | 5.33 | 10000 | 0.9038 | 64.3025 | | 0.8933 | 5.46 | 10250 | 0.9021 | 64.2022 | | 0.942 | 5.59 | 10500 | 0.8994 | 64.2339 | | 0.8833 | 5.73 | 10750 | 0.8966 | 63.8867 | | 0.9219 | 5.86 | 11000 | 0.8927 | 63.8603 | | 0.8513 | 5.99 | 11250 | 0.8918 | 64.0477 | | 0.9027 | 6.13 | 11500 | 0.8901 | 63.4577 | | 0.9134 | 6.26 | 11750 | 0.8874 | 63.3310 | | 0.9074 | 6.39 | 12000 | 0.8857 | 63.1198 | | 0.8905 | 6.53 | 12250 | 0.8834 | 63.0736 | | 0.8946 | 6.66 | 12500 | 0.8810 | 62.9455 | | 0.8554 | 6.79 | 12750 | 0.8780 | 62.8953 | | 0.8543 | 6.93 | 13000 | 0.8771 | 62.7884 | | 0.8646 | 7.06 | 13250 | 0.8757 | 62.6419 | | 0.8466 | 7.19 | 13500 | 0.8728 | 62.4848 | | 0.8781 | 7.33 | 13750 | 0.8717 | 62.5614 | | 0.9009 | 7.46 | 14000 | 0.8693 | 62.3858 | | 0.8451 | 7.59 | 14250 | 0.8687 | 62.1113 | | 0.8414 | 7.73 | 14500 | 0.8657 | 62.1403 | | 0.8444 | 7.86 | 14750 | 0.8650 | 61.9014 | | 0.875 | 7.99 | 15000 | 0.8636 | 61.7971 | | 0.7954 | 8.12 | 15250 | 0.8619 | 61.8076 | | 0.8818 | 8.26 | 15500 | 0.8597 | 61.5846 | | 0.8344 | 8.39 | 15750 | 0.8580 | 61.6413 | | 0.8402 | 8.52 | 16000 | 0.8567 | 61.6572 | | 0.8051 | 8.66 | 16250 | 0.8568 | 61.4948 | | 0.8563 | 8.79 | 16500 | 0.8544 | 61.3549 | | 0.8482 | 8.92 | 16750 | 0.8532 | 61.0024 | | 0.8283 | 9.06 | 17000 | 0.8521 | 61.0090 | | 0.8542 | 9.19 | 17250 | 0.8500 | 61.1199 | | 0.8433 | 9.32 | 17500 | 0.8486 | 61.0130 | | 0.7982 | 9.46 | 17750 | 0.8480 | 61.0183 | | 0.8372 | 9.59 | 18000 | 0.8477 | 60.8467 | | 0.8059 | 9.72 | 18250 | 0.8456 | 60.7727 | | 0.7764 | 9.86 | 18500 | 0.8458 | 60.5061 | | 0.8421 | 9.99 | 18750 | 0.8440 | 60.5259 | | 0.8334 | 10.12 | 19000 | 0.8436 | 60.7081 | | 0.7788 | 10.26 | 19250 | 0.8421 | 60.3253 | | 0.7844 | 10.39 | 19500 | 0.8411 | 60.2896 | | 0.7977 | 10.52 | 19750 | 0.8389 | 60.3833 | | 0.7955 | 10.66 | 20000 | 0.8385 | 60.3873 | | 0.7735 | 10.79 | 20250 | 0.8376 | 60.2355 | | 0.7384 | 10.92 | 20500 | 0.8367 | 60.3332 | | 0.7843 | 11.05 | 20750 | 0.8371 | 60.3926 | | 0.7242 | 11.19 | 21000 | 0.8358 | 60.2645 | | 0.7681 | 11.32 | 21250 | 0.8349 | 60.0876 | | 0.7691 | 11.45 | 21500 | 0.8347 | 60.1484 | | 0.7833 | 11.59 | 21750 | 0.8317 | 60.1906 | | 0.7346 | 11.72 | 22000 | 0.8325 | 59.9543 | | 0.78 | 11.85 | 22250 | 0.8309 | 59.8329 | | 0.7717 | 11.99 | 22500 | 0.8302 | 59.8764 | | 0.7368 | 12.12 | 22750 | 0.8284 | 59.9015 | | 0.7953 | 12.25 | 23000 | 0.8277 | 59.7708 | | 0.775 | 12.39 | 23250 | 0.8275 | 59.6626 | | 0.7301 | 12.52 | 23500 | 0.8268 | 59.5953 | | 0.7346 | 12.65 | 23750 | 0.8275 | 59.5332 | | 0.7411 | 12.79 | 24000 | 0.8266 | 59.4276 | | 0.7371 | 12.92 | 24250 | 0.8263 | 59.4936 | | 0.6947 | 13.05 | 24500 | 0.8262 | 59.4606 | | 0.775 | 13.19 | 24750 | 0.8229 | 59.3999 | | 0.7298 | 13.32 | 25000 | 0.8228 | 59.3907 | | 0.7356 | 13.45 | 25250 | 0.8233 | 59.1887 | | 0.69 | 13.59 | 25500 | 0.8233 | 59.2785 | | 0.7767 | 13.72 | 25750 | 0.8223 | 59.1055 | | 0.6991 | 13.85 | 26000 | 0.8221 | 59.0976 | | 0.7089 | 13.99 | 26250 | 0.8210 | 59.1148 | | 0.6992 | 14.12 | 26500 | 0.8207 | 59.0488 | | 0.7568 | 14.25 | 26750 | 0.8210 | 58.8429 | | 0.7371 | 14.38 | 27000 | 0.8206 | 58.8825 | | 0.7004 | 14.52 | 27250 | 0.8198 | 58.8415 | | 0.7151 | 14.65 | 27500 | 0.8174 | 58.7808 | | 0.7258 | 14.78 | 27750 | 0.8175 | 58.6818 | | 0.7447 | 14.92 | 28000 | 0.8181 | 58.8112 | | 0.6924 | 15.05 | 28250 | 0.8175 | 58.6977 | | 0.7266 | 15.18 | 28500 | 0.8170 | 58.7320 | | 0.6851 | 15.32 | 28750 | 0.8181 | 58.5551 | | 0.7061 | 15.45 | 29000 | 0.8175 | 58.4403 | | 0.693 | 15.58 | 29250 | 0.8176 | 58.4957 | | 0.7151 | 15.72 | 29500 | 0.8151 | 58.4785 | | 0.7142 | 15.85 | 29750 | 0.8148 | 58.2977 | | 0.7139 | 15.98 | 30000 | 0.8150 | 58.3175 | | 0.6746 | 16.12 | 30250 | 0.8172 | 58.2145 | | 0.7114 | 16.25 | 30500 | 0.8153 | 58.0390 | | 0.6772 | 16.38 | 30750 | 0.8142 | 58.0812 | | 0.6703 | 16.52 | 31000 | 0.8148 | 58.1855 | | 0.6746 | 16.65 | 31250 | 0.8126 | 58.2766 | | 0.7596 | 16.78 | 31500 | 0.8123 | 58.0152 | | 0.6705 | 16.92 | 31750 | 0.8119 | 58.0535 | | 0.678 | 17.05 | 32000 | 0.8126 | 58.2225 | | 0.6663 | 17.18 | 32250 | 0.8126 | 58.1221 | | 0.6555 | 17.31 | 32500 | 0.8124 | 57.9030 | | 0.6731 | 17.45 | 32750 | 0.8123 | 57.9967 | | 0.6533 | 17.58 | 33000 | 0.8117 | 57.9096 | | 0.667 | 17.71 | 33250 | 0.8107 | 57.7314 | | 0.6703 | 17.85 | 33500 | 0.8107 | 57.7103 | | 0.6481 | 17.98 | 33750 | 0.8121 | 57.6390 | | 0.6794 | 18.11 | 34000 | 0.8113 | 57.5902 | | 0.6871 | 18.25 | 34250 | 0.8099 | 57.6113 | | 0.7028 | 18.38 | 34500 | 0.8078 | 57.6390 | | 0.6626 | 18.51 | 34750 | 0.8110 | 57.7472 | | 0.6832 | 18.65 | 35000 | 0.8105 | 57.5136 | | 0.6468 | 18.78 | 35250 | 0.8073 | 57.6020 | | 0.6323 | 18.91 | 35500 | 0.8121 | 57.5677 | | 0.6435 | 19.05 | 35750 | 0.8090 | 57.5057 | | 0.6299 | 19.18 | 36000 | 0.8105 | 57.5849 | | 0.6829 | 19.31 | 36250 | 0.8083 | 57.5440 | | 0.6605 | 19.45 | 36500 | 0.8090 | 57.5176 | | 0.6429 | 19.58 | 36750 | 0.8091 | 57.3407 | | 0.6647 | 19.71 | 37000 | 0.8088 | 57.4106 | | 0.6279 | 19.85 | 37250 | 0.8108 | 57.4225 | | 0.6401 | 19.98 | 37500 | 0.8087 | 57.1664 | | 0.6361 | 20.11 | 37750 | 0.8082 | 57.0067 | | 0.6377 | 20.25 | 38000 | 0.8091 | 57.0225 | | 0.6439 | 20.38 | 38250 | 0.8102 | 57.1796 | | 0.6521 | 20.51 | 38500 | 0.8108 | 57.1374 | | 0.635 | 20.64 | 38750 | 0.8115 | 56.9592 | | 0.6313 | 20.78 | 39000 | 0.8097 | 56.9961 | | 0.6163 | 20.91 | 39250 | 0.8062 | 57.1163 | | 0.6242 | 21.04 | 39500 | 0.8084 | 56.9856 | | 0.5865 | 21.18 | 39750 | 0.8085 | 57.0001 | | 0.643 | 21.31 | 40000 | 0.8083 | 56.9288 | | 0.6226 | 21.44 | 40250 | 0.8103 | 56.8417 | | 0.5806 | 21.58 | 40500 | 0.8095 | 56.9513 | | 0.5979 | 21.71 | 40750 | 0.8103 | 56.7612 | | 0.5719 | 21.84 | 41000 | 0.8084 | 56.7678 | | 0.6497 | 21.98 | 41250 | 0.8083 | 56.8681 | | 0.6261 | 22.11 | 41500 | 0.8089 | 56.8140 | | 0.6303 | 22.24 | 41750 | 0.8091 | 56.8483 | | 0.547 | 22.38 | 42000 | 0.8078 | 56.7031 | | 0.6221 | 22.51 | 42250 | 0.8078 | 56.6595 | | 0.6047 | 22.64 | 42500 | 0.8078 | 56.7559 | | 0.5946 | 22.78 | 42750 | 0.8075 | 56.6688 | | 0.6289 | 22.91 | 43000 | 0.8082 | 56.6701 | | 0.6382 | 23.04 | 43250 | 0.8090 | 56.6688 | | 0.6002 | 23.18 | 43500 | 0.8104 | 56.6490 | | 0.5993 | 23.31 | 43750 | 0.8056 | 56.5658 | | 0.5435 | 23.44 | 44000 | 0.8092 | 56.6292 | | 0.5884 | 23.57 | 44250 | 0.8084 | 56.5315 | | 0.5689 | 23.71 | 44500 | 0.8102 | 56.6622 | | 0.5892 | 23.84 | 44750 | 0.8066 | 56.3850 | | 0.5634 | 23.97 | 45000 | 0.8106 | 56.4668 | | 0.611 | 24.11 | 45250 | 0.8098 | 56.4761 | | 0.5657 | 24.24 | 45500 | 0.8100 | 56.5421 | | 0.5877 | 24.37 | 45750 | 0.8100 | 56.3771 | | 0.6074 | 24.51 | 46000 | 0.8108 | 56.4272 | | 0.5609 | 24.64 | 46250 | 0.8120 | 56.5302 | | 0.5863 | 24.77 | 46500 | 0.8079 | 56.3969 | | 0.5547 | 24.91 | 46750 | 0.8109 | 56.2371 | ### Framework versions - Transformers 4.33.2 - Pytorch 2.0.1 - Datasets 2.19.1 - Tokenizers 0.13.3