Hugging Face's logo

cantillation
/
Teamim-large-v2_Random-True_OldData_date-07-06-2024_16-07-57

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Hebrew
whisper
hf-asr-leaderboard
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community

he-cantillation

This model is a fine-tuned version of Teamim-large-v2_Random-True_date-06-06-2024_21-59-43 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1326
  • Wer: 13.4139
  • Avg Precision Exact: 0.9006
  • Avg Recall Exact: 0.9003
  • Avg F1 Exact: 0.8998
  • Avg Precision Letter Shift: 0.9238
  • Avg Recall Letter Shift: 0.9240
  • Avg F1 Letter Shift: 0.9232
  • Avg Precision Word Level: 0.9256
  • Avg Recall Word Level: 0.9251
  • Avg F1 Word Level: 0.9246
  • Avg Precision Word Shift: 0.9608
  • Avg Recall Word Shift: 0.9621
  • Avg F1 Word Shift: 0.9608
  • Precision Median Exact: 0.9286
  • Recall Median Exact: 0.9286
  • F1 Median Exact: 0.9524
  • 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: 0.6364
  • Recall Min Word Shift: 0.5833
  • F1 Min Word Shift: 0.6087

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: 50
  • training_steps: 8000
  • 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
No log 8e-05 1 1.4938 84.8485 0.2550 0.2358 0.2443 0.2950 0.2723 0.2823 0.3128 0.2852 0.2975 0.4888 0.4600 0.4720 0.2308 0.2143 0.2222 0.8333 0.8333 0.8333 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.2356 0.08 1000 0.1521 18.4573 0.8447 0.8494 0.8462 0.8707 0.8762 0.8727 0.8751 0.8788 0.8761 0.9351 0.9437 0.9385 0.9167 0.9167 0.9167 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5385 0.5833 0.5600
0.1781 0.16 2000 0.1429 16.1263 0.8883 0.8917 0.8892 0.9148 0.9188 0.9159 0.9165 0.9196 0.9173 0.9495 0.9556 0.9517 0.9231 0.9231 0.9231 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6923 0.6667 0.6957
0.1448 0.24 3000 0.1324 15.5118 0.8804 0.8821 0.8804 0.9043 0.9065 0.9046 0.9083 0.9095 0.9081 0.9537 0.9579 0.9550 0.9231 0.9231 0.9286 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.5556 0.5263
0.1089 0.32 4000 0.1300 14.4522 0.8863 0.8888 0.8868 0.9069 0.9104 0.9079 0.9095 0.9134 0.9107 0.9530 0.9565 0.9539 0.9231 0.9231 0.9474 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6154 0.5 0.5517
0.0674 0.4 5000 0.1379 14.3463 0.8878 0.8942 0.8901 0.9122 0.9186 0.9146 0.9147 0.9203 0.9167 0.9563 0.9636 0.9591 0.9231 0.9286 0.9524 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6364 0.5833 0.6087
0.0548 0.48 6000 0.1380 13.8589 0.9005 0.9047 0.9018 0.9214 0.9262 0.9230 0.9239 0.9282 0.9253 0.9588 0.9650 0.9612 0.9286 0.9286 0.9524 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.5833 0.6087
0.0461 0.56 7000 0.1343 13.6046 0.9012 0.9072 0.9034 0.9246 0.9309 0.9269 0.9268 0.9322 0.9288 0.9590 0.9646 0.9611 0.9286 0.9333 0.9524 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6364 0.5833 0.6087
0.0608 0.64 8000 0.1326 13.4139 0.9006 0.9003 0.8998 0.9238 0.9240 0.9232 0.9256 0.9251 0.9246 0.9608 0.9621 0.9608 0.9286 0.9286 0.9524 1.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6364 0.5833 0.6087

Framework versions

  • Transformers 4.42.0.dev0
  • Pytorch 1.13.1+cu117
  • Datasets 2.16.1
  • Tokenizers 0.19.1
Downloads last month
23
Safetensors
Model size
1.54B params
Tensor type
F32
·
Inference Examples
Automatic Speech Recognition
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for cantillation/Teamim-large-v2_Random-True_OldData_date-07-06-2024_16-07-57

Finetunes
1 model

Evaluation results

Metadata error: specify a dataset to view leaderboard