--- language: - he license: apache-2.0 base_model: openai/whisper-medium tags: - hf-asr-leaderboard - generated_from_trainer metrics: - wer 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: 0.1692 - Wer: 21.1641 - Avg Precision Exact: 0.8062 - Avg Recall Exact: 0.8080 - Avg F1 Exact: 0.8065 - Avg Precision Letter Shift: 0.8352 - Avg Recall Letter Shift: 0.8370 - Avg F1 Letter Shift: 0.8355 - Avg Precision Word Level: 0.8409 - Avg Recall Word Level: 0.8430 - Avg F1 Word Level: 0.8413 - Avg Precision Word Shift: 0.9407 - Avg Recall Word Shift: 0.9446 - Avg F1 Word Shift: 0.9419 - Precision Median Exact: 0.9091 - Recall Median Exact: 0.9091 - F1 Median Exact: 0.9091 - Precision Max Exact: 1.0 - Recall Max Exact: 1.0 - F1 Max Exact: 1.0 - Precision Min Exact: 0.0 - Recall Min Exact: 0.0 - F1 Min Exact: 0.0 - Precision Min Letter Shift: 0.0 - Recall Min Letter Shift: 0.0 - F1 Min Letter Shift: 0.0 - Precision Min Word Level: 0.0 - Recall Min Word Level: 0.0 - F1 Min Word Level: 0.0 - Precision Min Word Shift: 0.1429 - Recall Min Word Shift: 0.1 - F1 Min Word Shift: 0.1176 ## 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: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 100 - training_steps: 30000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | Avg Precision Exact | Avg Recall Exact | Avg F1 Exact | Avg Precision Letter Shift | Avg Recall Letter Shift | Avg F1 Letter Shift | Avg Precision Word Level | Avg Recall Word Level | Avg F1 Word Level | Avg Precision Word Shift | Avg Recall Word Shift | Avg F1 Word Shift | Precision Median Exact | Recall Median Exact | F1 Median Exact | Precision Max Exact | Recall Max Exact | F1 Max Exact | Precision Min Exact | Recall Min Exact | F1 Min Exact | Precision Min Letter Shift | Recall Min Letter Shift | F1 Min Letter Shift | Precision Min Word Level | Recall Min Word Level | F1 Min Word Level | Precision Min Word Shift | Recall Min Word Shift | F1 Min Word Shift | |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------------------:|:----------------:|:------------:|:--------------------------:|:-----------------------:|:-------------------:|:------------------------:|:---------------------:|:-----------------:|:------------------------:|:---------------------:|:-----------------:|:----------------------:|:-------------------:|:---------------:|:-------------------:|:----------------:|:------------:|:-------------------:|:----------------:|:------------:|:--------------------------:|:-----------------------:|:-------------------:|:------------------------:|:---------------------:|:-----------------:|:------------------------:|:---------------------:|:-----------------:| | 0.2936 | 0.16 | 2000 | 0.3109 | 44.8004 | 0.5576 | 0.5646 | 0.5602 | 0.6021 | 0.6096 | 0.6049 | 0.6142 | 0.6239 | 0.6181 | 0.8035 | 0.8191 | 0.8099 | 0.6154 | 0.625 | 0.6207 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.1696 | 0.32 | 4000 | 0.2363 | 35.3622 | 0.6570 | 0.6509 | 0.6531 | 0.6965 | 0.6898 | 0.6923 | 0.7059 | 0.7000 | 0.7021 | 0.8774 | 0.8761 | 0.8755 | 0.75 | 0.75 | 0.75 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.1147 | 0.48 | 6000 | 0.2059 | 30.2846 | 0.7069 | 0.7015 | 0.7035 | 0.7422 | 0.7364 | 0.7385 | 0.7512 | 0.7461 | 0.7478 | 0.8992 | 0.8989 | 0.8980 | 0.8182 | 0.8 | 0.8000 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.1 | 0.1176 | | 0.101 | 0.64 | 8000 | 0.1887 | 27.1175 | 0.7477 | 0.7487 | 0.7475 | 0.7812 | 0.7822 | 0.7810 | 0.7884 | 0.7898 | 0.7884 | 0.9171 | 0.9216 | 0.9183 | 0.8462 | 0.8571 | 0.8571 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0625 | 0.8 | 10000 | 0.1856 | 25.9313 | 0.7457 | 0.7570 | 0.7506 | 0.7776 | 0.7894 | 0.7827 | 0.7838 | 0.7952 | 0.7888 | 0.9171 | 0.9287 | 0.9219 | 0.8571 | 0.875 | 0.8571 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.1 | 0.1176 | | 0.0744 | 0.96 | 12000 | 0.1771 | 24.8226 | 0.7654 | 0.7722 | 0.7681 | 0.7948 | 0.8019 | 0.7976 | 0.8016 | 0.8085 | 0.8043 | 0.9189 | 0.9265 | 0.9218 | 0.8667 | 0.875 | 0.8696 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0769 | 0.0909 | 0.0833 | | 0.0596 | 1.12 | 14000 | 0.1725 | 23.7103 | 0.7773 | 0.7794 | 0.7777 | 0.8077 | 0.8100 | 0.8082 | 0.8148 | 0.8169 | 0.8151 | 0.9306 | 0.9334 | 0.9311 | 0.8889 | 0.8889 | 0.8889 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.1 | 0.1176 | | 0.0515 | 1.28 | 16000 | 0.1696 | 22.9305 | 0.7880 | 0.7897 | 0.7883 | 0.8183 | 0.8200 | 0.8185 | 0.8242 | 0.8261 | 0.8245 | 0.9352 | 0.9384 | 0.9360 | 0.9 | 0.9 | 0.8966 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1111 | 0.1 | 0.1176 | | 0.0369 | 1.44 | 18000 | 0.1695 | 22.4390 | 0.7937 | 0.7924 | 0.7925 | 0.8239 | 0.8226 | 0.8226 | 0.8297 | 0.8289 | 0.8286 | 0.9370 | 0.9391 | 0.9372 | 0.9 | 0.9 | 0.9 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.035 | 1.6 | 20000 | 0.1699 | 22.2358 | 0.7934 | 0.7948 | 0.7935 | 0.8227 | 0.8242 | 0.8228 | 0.8288 | 0.8303 | 0.8289 | 0.9380 | 0.9412 | 0.9388 | 0.9 | 0.9 | 0.9 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | | 0.0282 | 1.76 | 22000 | 0.1686 | 21.9549 | 0.7956 | 0.7940 | 0.7942 | 0.8258 | 0.8241 | 0.8243 | 0.8314 | 0.8298 | 0.8300 | 0.9408 | 0.9412 | 0.9402 | 0.9 | 0.9 | 0.9 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1111 | 0.1 | 0.1176 | | 0.0215 | 1.92 | 24000 | 0.1688 | 21.6445 | 0.8002 | 0.8022 | 0.8006 | 0.8287 | 0.8307 | 0.8291 | 0.8341 | 0.8363 | 0.8346 | 0.9380 | 0.9417 | 0.9391 | 0.9 | 0.9091 | 0.9 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.1 | 0.1176 | | 0.024 | 2.08 | 26000 | 0.1699 | 21.1899 | 0.8037 | 0.8070 | 0.8047 | 0.8331 | 0.8365 | 0.8342 | 0.8389 | 0.8424 | 0.8400 | 0.9415 | 0.9468 | 0.9434 | 0.9091 | 0.9091 | 0.9032 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.1 | 0.1176 | | 0.0198 | 2.24 | 28000 | 0.1696 | 21.3119 | 0.8038 | 0.8070 | 0.8048 | 0.8327 | 0.8360 | 0.8337 | 0.8382 | 0.8418 | 0.8393 | 0.9390 | 0.9445 | 0.9409 | 0.9091 | 0.9091 | 0.9032 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.1 | 0.1176 | | 0.0219 | 2.4 | 30000 | 0.1692 | 21.1641 | 0.8062 | 0.8080 | 0.8065 | 0.8352 | 0.8370 | 0.8355 | 0.8409 | 0.8430 | 0.8413 | 0.9407 | 0.9446 | 0.9419 | 0.9091 | 0.9091 | 0.9091 | 1.0 | 1.0 | 1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1429 | 0.1 | 0.1176 | ### Framework versions - Transformers 4.39.0.dev0 - Pytorch 2.2.1+cu121 - Datasets 2.16.1 - Tokenizers 0.15.0