You can deploy this model and start earning money today!

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

Name: Teamim-large-v2_Random-True_OldData_date-07-06-2024_16-07-57
Availability: InStock

automatic-speech-recognition

hf-asr-leaderboard

generated_from_trainer

endpoints_compatible

Views

No views yet

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 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
0.2356	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	0.5385	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	0.6923	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	0.5	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	0.6154	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	0.6364	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	0.6	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	0.6364	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	0.6364	0.5833	0.6087

Framework versions

Transformers 4.42.0.dev0
Pytorch 1.13.1+cu117
Datasets 2.16.1
Tokenizers 0.19.1

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

he-cantillation

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

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