Teamim-large-v2_Random-True_DropOut-0.3_Augmented_date-27-06-2024_20-55-08

cantillation/Teamim-large-v2_Random-True_DropOut-0.3_Augmented_date-27-06-2024_20-55-08

Teamim-large-v2_Random-True_DropOut-0.3_Augmented_date-27-06-2024_20-55-08 is a automatic speech recognition model from cantillation. Use it when you need speech turned into text. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

13% of all-time downloads

All-time downloads

Public

Parameters

1.5B

6.2 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors6.2 GB · 100%

At a glance

Task: Automatic Speech Recognition
Library: transformers
Model type: whisper
Access: Public
Created: Jun 27, 2024
Updated: Jun 28, 2024
SHA: d5be31eb

Base models

ivrit-ai/whisper-v2-pd1-e1

Model card

From the Hugging Face model README

he-cantillation

This model is a fine-tuned version of ivrit-ai/whisper-v2-pd1-e1 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.1065
Wer: 11.5595
Avg Precision Exact: 0.9158
Avg Recall Exact: 0.9153
Avg F1 Exact: 0.9152
Avg Precision Letter Shift: 0.9349
Avg Recall Letter Shift: 0.9345
Avg F1 Letter Shift: 0.9344
Avg Precision Word Level: 0.9374
Avg Recall Word Level: 0.9371
Avg F1 Word Level: 0.9369
Avg Precision Word Shift: 0.9744
Avg Recall Word Shift: 0.9751
Avg F1 Word Shift: 0.9743
Precision Median Exact: 1.0
Recall Median Exact: 1.0
F1 Median Exact: 1.0
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: 0.2222
Recall Min Word Shift: 0.1667
F1 Min Word Shift: 0.1905

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: 10000
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	4.8435	119.6859	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	0.0	0.0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
0.1248	0.08	1000	0.1621	23.2816	0.8165	0.8201	0.8175	0.8444	0.8482	0.8454	0.8487	0.8532	0.8501	0.9200	0.9275	0.9227	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	0.0909	0.1111
0.0907	0.16	2000	0.1268	18.2188	0.8624	0.8589	0.8600	0.8863	0.8828	0.8839	0.8905	0.8869	0.8880	0.9507	0.9495	0.9494	0.9231	0.9167	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	0.1429	0.125	0.1333
0.04	0.24	3000	0.1182	14.9446	0.8826	0.8813	0.8815	0.9046	0.9035	0.9035	0.9079	0.9067	0.9068	0.9633	0.9635	0.9628	0.9375	0.9375	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	0.125	0.1	0.1176
0.0379	0.32	4000	0.1123	15.0111	0.8839	0.8853	0.8839	0.9081	0.9096	0.9082	0.9110	0.9134	0.9115	0.9615	0.9658	0.9629	0.9333	0.9333	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	0.1	0.1	0.1176
0.0274	0.4	5000	0.1092	13.3481	0.9011	0.8989	0.8996	0.9215	0.9194	0.9200	0.9240	0.9225	0.9228	0.9684	0.9685	0.9679	1.0	1.0	0.9630	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
0.0242	0.48	6000	0.1078	13.0044	0.9003	0.8977	0.8985	0.9211	0.9187	0.9194	0.9237	0.9216	0.9221	0.9686	0.9681	0.9678	1.0	1.0	0.9630	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
0.0134	0.56	7000	0.1039	12.0547	0.9086	0.9072	0.9075	0.9270	0.9257	0.9259	0.9299	0.9288	0.9290	0.9699	0.9701	0.9696	1.0	1.0	1.0	1.0	1.0	1.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.1538	0.1429	0.1538
0.0118	0.64	8000	0.1077	11.8847	0.9104	0.9110	0.9103	0.9306	0.9313	0.9306	0.9331	0.9339	0.9331	0.9726	0.9748	0.9733	1.0	1.0	1.0	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
0.0084	0.72	9000	0.1071	11.6223	0.9145	0.9141	0.9140	0.9336	0.9332	0.9330	0.9358	0.9355	0.9353	0.9742	0.9751	0.9742	1.0	1.0	1.0	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
0.0065	0.8	10000	0.1065	11.5595	0.9158	0.9153	0.9152	0.9349	0.9345	0.9344	0.9374	0.9371	0.9369	0.9744	0.9751	0.9743	1.0	1.0	1.0	1.0	1.0	1.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.2222	0.1667	0.1905

Framework versions

Transformers 4.41.2
Pytorch 2.2.1
Datasets 2.20.0
Tokenizers 0.19.1

Task: Automatic Speech Recognition
Library: transformers
Type: whisper
Languages: he
Created: Jun 27, 2024
Updated: Jun 28, 2024