--- language: - he license: apache-2.0 base_model: openai/whisper-medium tags: - hf-asr-leaderboard - generated_from_trainer metrics: - precision - recall - f1 model-index: - name: he results: [] --- # he This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 1.6591 - Precision: 0.0398 - Recall: 0.0391 - F1: 0.0391 - Precision Median: 0.0 - Recall Median: 0.0 - F1 Median: 0.0 - Precision Max: 0.2 - Recall Max: 0.2 - F1 Max: 0.2000 - Precision Min: 0.0 - Recall Min: 0.0 - F1 Min: 0.0 ## 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: 0.0002 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1 - training_steps: 1000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Precision Median | Recall Median | F1 Median | Precision Max | Recall Max | F1 Max | Precision Min | Recall Min | F1 Min | |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:----------------:|:-------------:|:---------:|:-------------:|:----------:|:------:|:-------------:|:----------:|:------:| | 7.5196 | 0.01 | 25 | 3.8632 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0 | 0 | | 3.5602 | 0.02 | 50 | 3.2631 | 0.0160 | 0.0368 | 0.0222 | 0.0303 | 0.0556 | 0.04 | 0.0312 | 0.0909 | 0.0455 | 0.0 | 0.0 | 0.0 | | 3.1184 | 0.03 | 75 | 3.0632 | 0.0044 | 0.0090 | 0.0059 | 0.0 | 0.0 | 0.0 | 0.0417 | 0.1 | 0.0588 | 0.0 | 0.0 | 0.0 | | 2.9519 | 0.04 | 100 | 3.0036 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0 | 0 | | 2.8338 | 0.05 | 125 | 2.7501 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0 | 0 | | 2.6873 | 0.06 | 150 | 2.6214 | 0.0758 | 0.0563 | 0.0643 | 0.0909 | 0.0625 | 0.0714 | 0.2727 | 0.2 | 0.2308 | 0.0 | 0.0 | 0.0 | | 2.5495 | 0.07 | 175 | 2.4574 | 0.0310 | 0.0374 | 0.0335 | 0.0 | 0.0 | 0.0 | 0.1176 | 0.2 | 0.1481 | 0.0 | 0.0 | 0.0 | | 2.4591 | 0.08 | 200 | 2.4156 | 0.0376 | 0.0571 | 0.0449 | 0.0476 | 0.0625 | 0.0541 | 0.0952 | 0.2 | 0.1290 | 0.0 | 0.0 | 0.0 | | 2.3637 | 0.09 | 225 | 2.4155 | 0.0197 | 0.0192 | 0.0194 | 0.0 | 0.0 | 0.0 | 0.125 | 0.125 | 0.125 | 0.0 | 0.0 | 0.0 | | 2.3093 | 0.1 | 250 | 2.2890 | 0.0395 | 0.0434 | 0.0410 | 0.0 | 0.0 | 0.0 | 0.125 | 0.1818 | 0.1481 | 0.0 | 0.0 | 0.0 | | 2.2333 | 0.11 | 275 | 2.3871 | 0.0817 | 0.0815 | 0.0813 | 0.0625 | 0.0714 | 0.0645 | 0.1875 | 0.1765 | 0.1818 | 0.0 | 0.0 | 0.0 | | 2.1783 | 0.12 | 300 | 2.2399 | 0.0188 | 0.0169 | 0.0176 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.1176 | 0.1290 | 0.0 | 0.0 | 0.0 | | 2.1766 | 0.13 | 325 | 2.1678 | 0.1201 | 0.1392 | 0.1284 | 0.1111 | 0.1333 | 0.1212 | 0.2778 | 0.2941 | 0.2857 | 0.0 | 0.0 | 0.0 | | 2.0944 | 0.14 | 350 | 2.2700 | 0.0088 | 0.0062 | 0.0073 | 0.0 | 0.0 | 0.0 | 0.0833 | 0.0625 | 0.0714 | 0.0 | 0.0 | 0.0 | | 2.0261 | 0.15 | 375 | 2.1503 | 0.0386 | 0.0381 | 0.0382 | 0.0 | 0.0 | 0.0 | 0.1333 | 0.125 | 0.1290 | 0.0 | 0.0 | 0.0 | | 1.9888 | 0.16 | 400 | 2.1727 | 0.0117 | 0.0121 | 0.0119 | 0.0 | 0.0 | 0.0 | 0.0556 | 0.0625 | 0.0588 | 0.0 | 0.0 | 0.0 | | 1.955 | 0.17 | 425 | 2.1425 | 0.0795 | 0.0947 | 0.0861 | 0.1053 | 0.1176 | 0.1143 | 0.1579 | 0.2 | 0.1765 | 0.0 | 0.0 | 0.0 | | 1.9361 | 0.18 | 450 | 2.0521 | 0.0489 | 0.0595 | 0.0533 | 0.0526 | 0.0588 | 0.0556 | 0.1333 | 0.1667 | 0.1290 | 0.0 | 0.0 | 0.0 | | 1.8902 | 0.19 | 475 | 2.0196 | 0.0066 | 0.0060 | 0.0063 | 0.0 | 0.0 | 0.0 | 0.0667 | 0.0588 | 0.0606 | 0.0 | 0.0 | 0.0 | | 1.8809 | 0.2 | 500 | 2.0199 | 0.0259 | 0.0293 | 0.0274 | 0.0 | 0.0 | 0.0 | 0.1111 | 0.125 | 0.1176 | 0.0 | 0.0 | 0.0 | | 1.837 | 0.21 | 525 | 1.9694 | 0.0066 | 0.0067 | 0.0066 | 0.0 | 0.0 | 0.0 | 0.0667 | 0.0714 | 0.0645 | 0.0 | 0.0 | 0.0 | | 1.8329 | 0.22 | 550 | 1.9790 | 0.0095 | 0.0106 | 0.0100 | 0.0 | 0.0 | 0.0 | 0.0625 | 0.0714 | 0.0645 | 0.0 | 0.0 | 0.0 | | 1.8233 | 0.23 | 575 | 1.8902 | 0.0930 | 0.1220 | 0.1049 | 0.0625 | 0.0833 | 0.0714 | 0.25 | 0.3333 | 0.2703 | 0.0 | 0.0 | 0.0 | | 1.7814 | 0.24 | 600 | 1.8473 | 0.0785 | 0.0939 | 0.0851 | 0.0588 | 0.0833 | 0.0690 | 0.2778 | 0.2941 | 0.2857 | 0.0 | 0.0 | 0.0 | | 1.7461 | 0.25 | 625 | 1.7672 | 0.0799 | 0.0924 | 0.0852 | 0.0588 | 0.0909 | 0.0714 | 0.1667 | 0.2143 | 0.1875 | 0.0 | 0.0 | 0.0 | | 1.7398 | 0.26 | 650 | 1.8822 | 0.0837 | 0.0902 | 0.0863 | 0.0625 | 0.0625 | 0.0625 | 0.3333 | 0.3125 | 0.3226 | 0.0 | 0.0 | 0.0 | | 1.7282 | 0.27 | 675 | 1.7915 | 0.0539 | 0.0704 | 0.0605 | 0.0455 | 0.0625 | 0.0526 | 0.1364 | 0.2 | 0.1622 | 0.0 | 0.0 | 0.0 | | 1.6463 | 0.28 | 700 | 1.8641 | 0.0292 | 0.0305 | 0.0298 | 0.0 | 0.0 | 0.0 | 0.125 | 0.1333 | 0.1290 | 0.0 | 0.0 | 0.0 | | 1.6577 | 0.29 | 725 | 1.7592 | 0.0803 | 0.0886 | 0.0838 | 0.0588 | 0.0714 | 0.0588 | 0.2353 | 0.2353 | 0.2353 | 0.0 | 0.0 | 0.0 | | 1.6675 | 0.3 | 750 | 1.7720 | 0.0316 | 0.0426 | 0.0359 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.2143 | 0.1714 | 0.0 | 0.0 | 0.0 | | 1.6416 | 0.31 | 775 | 1.7664 | 0.0560 | 0.0674 | 0.0606 | 0.05 | 0.0625 | 0.0588 | 0.1667 | 0.1818 | 0.1714 | 0.0 | 0.0 | 0.0 | | 1.6431 | 0.32 | 800 | 1.7466 | 0.0778 | 0.1102 | 0.0893 | 0.0714 | 0.0909 | 0.0714 | 0.25 | 0.3636 | 0.2778 | 0.0 | 0.0 | 0.0 | | 1.5706 | 0.33 | 825 | 1.7675 | 0.0603 | 0.0646 | 0.0621 | 0.0526 | 0.0625 | 0.0588 | 0.1579 | 0.1875 | 0.1714 | 0.0 | 0.0 | 0.0 | | 1.5303 | 0.34 | 850 | 1.7940 | 0.0133 | 0.0162 | 0.0146 | 0.0 | 0.0 | 0.0 | 0.0556 | 0.0714 | 0.0588 | 0.0 | 0.0 | 0.0 | | 1.5507 | 0.35 | 875 | 1.7630 | 0.0371 | 0.0306 | 0.0334 | 0.0 | 0.0 | 0.0 | 0.1667 | 0.1429 | 0.1538 | 0.0 | 0.0 | 0.0 | | 1.5245 | 0.36 | 900 | 1.7160 | 0.0526 | 0.0409 | 0.0456 | 0.0769 | 0.05 | 0.0645 | 0.2 | 0.1818 | 0.1905 | 0.0 | 0.0 | 0.0 | | 1.5263 | 0.37 | 925 | 1.6858 | 0.0418 | 0.0429 | 0.0420 | 0.0 | 0.0 | 0.0 | 0.1333 | 0.1538 | 0.1333 | 0.0 | 0.0 | 0.0 | | 1.517 | 0.38 | 950 | 1.6739 | 0.0360 | 0.0380 | 0.0368 | 0.0 | 0.0 | 0.0 | 0.2 | 0.2 | 0.2000 | 0.0 | 0.0 | 0.0 | | 1.488 | 0.39 | 975 | 1.6761 | 0.0395 | 0.0461 | 0.0425 | 0.05 | 0.0556 | 0.0526 | 0.1176 | 0.1429 | 0.1290 | 0.0 | 0.0 | 0.0 | | 1.4926 | 0.4 | 1000 | 1.6591 | 0.0398 | 0.0391 | 0.0391 | 0.0 | 0.0 | 0.0 | 0.2 | 0.2 | 0.2000 | 0.0 | 0.0 | 0.0 | ### Framework versions - Transformers 4.36.2 - Pytorch 1.13.1+cu117 - Datasets 2.16.1 - Tokenizers 0.15.0