--- 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: 0.1756 - Wer: 20.1811 - Avg Precision Exact: 0.8083 - Avg Recall Exact: 0.8102 - Avg F1 Exact: 0.8087 - Avg Precision Letter Shift: 0.8373 - Avg Recall Letter Shift: 0.8394 - Avg F1 Letter Shift: 0.8377 - Avg Precision Word Level: 0.8427 - Avg Recall Word Level: 0.8450 - Avg F1 Word Level: 0.8432 - Avg Precision Word Shift: 0.9448 - Avg Recall Word Shift: 0.9489 - Avg F1 Word Shift: 0.9460 - 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: 40000 - 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 | |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------------------:|:----------------:|:------------:|:--------------------------:|:-----------------------:|:-------------------:|:------------------------:|:---------------------:|:-----------------:|:------------------------:|:---------------------:|:-----------------:|:----------------------:|:-------------------:|:---------------:|:-------------------:|:----------------:|:------------:|:-------------------:|:----------------:|:------------:|:--------------------------:|:-----------------------:|:-------------------:|:------------------------:|:---------------------:|:-----------------:|:------------------------:|:---------------------:|:-----------------:| | 0.2873 | 0.16 | 2000 | 0.3088 | 44.3311 | 0.5598 | 0.5685 | 0.5633 | 0.6026 | 0.6118 | 0.6062 | 0.6138 | 0.6239 | 0.6178 | 0.8019 | 0.8196 | 0.8092 | 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 | | 0.1752 | 0.32 | 4000 | 0.2328 | 35.1811 | 0.6557 | 0.6595 | 0.6568 | 0.6946 | 0.6985 | 0.6957 | 0.7041 | 0.7082 | 0.7053 | 0.8676 | 0.8745 | 0.8698 | 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 | | 0.117 | 0.48 | 6000 | 0.1997 | 29.4605 | 0.7124 | 0.7125 | 0.7117 | 0.7514 | 0.7513 | 0.7506 | 0.7604 | 0.7606 | 0.7597 | 0.9031 | 0.9063 | 0.9037 | 0.8182 | 0.8182 | 0.8148 | 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 | 0.1053 | | 0.0994 | 0.64 | 8000 | 0.1881 | 27.5610 | 0.7359 | 0.7407 | 0.7376 | 0.7708 | 0.7758 | 0.7726 | 0.7783 | 0.7837 | 0.7803 | 0.9117 | 0.9191 | 0.9144 | 0.8333 | 0.8462 | 0.8387 | 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 | 0.0909 | 0.1111 | | 0.0664 | 0.8 | 10000 | 0.1837 | 25.9682 | 0.7446 | 0.7529 | 0.7480 | 0.7785 | 0.7873 | 0.7821 | 0.7857 | 0.7944 | 0.7893 | 0.9194 | 0.9277 | 0.9226 | 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 | 0.1111 | 0.0909 | 0.1111 | | 0.0767 | 0.96 | 12000 | 0.1760 | 24.6157 | 0.7561 | 0.7662 | 0.7604 | 0.7868 | 0.7973 | 0.7913 | 0.7936 | 0.8040 | 0.7980 | 0.9194 | 0.9315 | 0.9245 | 0.8667 | 0.875 | 0.8723 | 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 | | 0.0593 | 1.12 | 14000 | 0.1716 | 23.6511 | 0.7669 | 0.7732 | 0.7694 | 0.7988 | 0.8054 | 0.8014 | 0.8047 | 0.8114 | 0.8073 | 0.9275 | 0.9343 | 0.9300 | 0.875 | 0.8889 | 0.8800 | 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 | 0.1111 | 0.1111 | | 0.0525 | 1.28 | 16000 | 0.1712 | 23.1264 | 0.7778 | 0.7788 | 0.7777 | 0.8092 | 0.8104 | 0.8092 | 0.8156 | 0.8169 | 0.8156 | 0.9345 | 0.9383 | 0.9356 | 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 | 0.1111 | 0.1111 | 0.125 | | 0.0358 | 1.44 | 18000 | 0.1699 | 22.4538 | 0.7841 | 0.7845 | 0.7837 | 0.8150 | 0.8157 | 0.8147 | 0.8212 | 0.8222 | 0.8211 | 0.9344 | 0.9376 | 0.9351 | 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 | 0.1 | 0.1 | 0.1053 | | 0.0323 | 1.6 | 20000 | 0.1713 | 22.2173 | 0.7873 | 0.7926 | 0.7893 | 0.8170 | 0.8224 | 0.8190 | 0.8230 | 0.8286 | 0.8251 | 0.9362 | 0.9424 | 0.9385 | 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 | 0.1111 | | 0.0248 | 1.76 | 22000 | 0.1683 | 21.7480 | 0.7934 | 0.7945 | 0.7933 | 0.8235 | 0.8248 | 0.8235 | 0.8294 | 0.8310 | 0.8295 | 0.9407 | 0.9433 | 0.9412 | 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 | 0.1429 | 0.1 | 0.1176 | | 0.0209 | 1.92 | 24000 | 0.1696 | 21.0310 | 0.7982 | 0.8000 | 0.7986 | 0.8275 | 0.8292 | 0.8278 | 0.8331 | 0.8351 | 0.8335 | 0.9424 | 0.9461 | 0.9435 | 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 | 0.1429 | 0.1 | 0.1176 | | 0.0202 | 2.08 | 26000 | 0.1713 | 21.1936 | 0.7954 | 0.7988 | 0.7965 | 0.8250 | 0.8285 | 0.8262 | 0.8309 | 0.8346 | 0.8321 | 0.9404 | 0.9460 | 0.9424 | 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 | 0.0769 | 0.0909 | 0.0833 | | 0.0172 | 2.24 | 28000 | 0.1716 | 20.7761 | 0.8013 | 0.8053 | 0.8027 | 0.8304 | 0.8346 | 0.8319 | 0.8359 | 0.8407 | 0.8376 | 0.9404 | 0.9469 | 0.9428 | 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 | 0.2 | 0.1818 | 0.2000 | | 0.0161 | 2.4 | 30000 | 0.1740 | 20.6135 | 0.8052 | 0.8079 | 0.8059 | 0.8351 | 0.8380 | 0.8359 | 0.8408 | 0.8440 | 0.8417 | 0.9439 | 0.9494 | 0.9459 | 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 | 0.125 | 0.1 | 0.1176 | | 0.01 | 2.56 | 32000 | 0.1743 | 20.6948 | 0.8031 | 0.8048 | 0.8033 | 0.8322 | 0.8339 | 0.8323 | 0.8380 | 0.8399 | 0.8382 | 0.9441 | 0.9480 | 0.9452 | 0.9091 | 0.9091 | 0.9062 | 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 | | 0.025 | 2.72 | 34000 | 0.1753 | 20.6282 | 0.8033 | 0.8072 | 0.8046 | 0.8327 | 0.8368 | 0.8341 | 0.8383 | 0.8430 | 0.8400 | 0.9419 | 0.9489 | 0.9446 | 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 | | 0.0082 | 2.88 | 36000 | 0.1756 | 20.3991 | 0.8060 | 0.8081 | 0.8064 | 0.8354 | 0.8378 | 0.8359 | 0.8406 | 0.8433 | 0.8412 | 0.9436 | 0.9484 | 0.9452 | 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 | | 0.013 | 3.04 | 38000 | 0.1754 | 20.3030 | 0.8078 | 0.8097 | 0.8082 | 0.8374 | 0.8395 | 0.8378 | 0.8427 | 0.8452 | 0.8433 | 0.9447 | 0.9488 | 0.9459 | 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 | | 0.0183 | 3.2 | 40000 | 0.1756 | 20.1811 | 0.8083 | 0.8102 | 0.8087 | 0.8373 | 0.8394 | 0.8377 | 0.8427 | 0.8450 | 0.8432 | 0.9448 | 0.9489 | 0.9460 | 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