Teamim-tiny_WeightDecay-0.05_Combined-Data_date-17-07-2024_10-10

cantillation/Teamim-tiny_WeightDecay-0.05_Combined-Data_date-17-07-2024_10-10

Teamim-tiny_WeightDecay-0.05_Combined-Data_date-17-07-2024_10-10 is a automatic speech recognition model from cantillation. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.

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

Downloads · 30 days

1% of all-time downloads

All-time downloads

343

Public

Parameters

37.8M

7.7 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors151 MB · 96%

At a glance

Task: Automatic Speech Recognition
Library: transformers
License: apache-2.0
Model type: whisper
Access: Public
Created: Jul 17, 2024
Updated: Jul 18, 2024
SHA: 2e0e2fad

Base models

openai/whisper-tiny

Model card

From the Hugging Face model README

he-cantillation

This model is a fine-tuned version of openai/whisper-tiny on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2642
Wer: 16.6210
Avg Precision Exact: 0.8369
Avg Recall Exact: 0.8382
Avg F1 Exact: 0.8371
Avg Precision Letter Shift: 0.8586
Avg Recall Letter Shift: 0.8599
Avg F1 Letter Shift: 0.8588
Avg Precision Word Level: 0.8633
Avg Recall Word Level: 0.8652
Avg F1 Word Level: 0.8637
Avg Precision Word Shift: 0.9480
Avg Recall Word Shift: 0.9510
Avg F1 Word Shift: 0.9488
Precision Median Exact: 0.9231
Recall Median Exact: 0.9231
F1 Median Exact: 0.9286
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: 0.125
F1 Min Word Shift: 0.1333

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: 500
training_steps: 500000
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	0.0001	1	9.1305	104.1002	0.0008	0.0017	0.0007	0.0055	0.0068	0.0048	0.0027	0.0187	0.0046	0.0242	0.0322	0.0233	0.0	0.0	0.0	0.1429	1.0	0.1818	0.0	0.0	0.0
0.1404	0.5167	10000	0.2062	31.3792	0.7018	0.7031	0.7017	0.7345	0.7359	0.7344	0.7437	0.7448	0.7434	0.8802	0.8850	0.8815	0.8	0.8	0.8	1.0	1.0	1.0	0.0769	0.125	0.1111
0.0565	1.0334	20000	0.1776	25.7434	0.7505	0.7534	0.7513	0.7792	0.7821	0.7799	0.7865	0.7900	0.7875	0.9069	0.9122	0.9086	0.8462	0.8462	0.8485	1.0	1.0	1.0	0.0769	0.0909	0.0833
0.0343	1.5501	30000	0.1768	24.2487	0.7700	0.7691	0.7689	0.7984	0.7974	0.7972	0.8052	0.8043	0.8040	0.9143	0.9158	0.9142	0.875	0.875	0.875	1.0	1.0	1.0	0.0	0.0	0.0
0.0202	2.0668	40000	0.1824	22.5935	0.7783	0.7853	0.7812	0.8052	0.8126	0.8082	0.8107	0.8185	0.8139	0.9196	0.9301	0.9240	0.8889	0.9	0.8889	1.0	1.0	1.0	0.1	0.1	0.1000
0.0136	2.5834	50000	0.1848	21.7408	0.7886	0.7912	0.7893	0.8142	0.8169	0.8150	0.8199	0.8229	0.8208	0.9234	0.9281	0.9249	0.9	0.9	0.9	1.0	1.0	1.0	0.1	0.1	0.1000
0.0338	3.1001	60000	0.1936	21.6149	0.7881	0.7909	0.7889	0.8137	0.8166	0.8146	0.8188	0.8233	0.8203	0.9222	0.9295	0.9250	0.9	0.9	0.9	1.0	1.0	1.0	0.1	0.1	0.1000
0.0148	3.6168	70000	0.1932	20.5198	0.8045	0.8086	0.8059	0.8295	0.8339	0.8311	0.8346	0.8394	0.8364	0.9261	0.9333	0.9289	0.9	0.9091	0.9	1.0	1.0	1.0	0.1	0.1	0.1000
0.007	4.1335	80000	0.1995	20.3562	0.8039	0.8048	0.8038	0.8285	0.8296	0.8285	0.8340	0.8356	0.8342	0.9314	0.9354	0.9326	0.9091	0.9091	0.9091	1.0	1.0	1.0	0.1429	0.1111	0.125
0.0073	4.6502	90000	0.2057	20.7275	0.8005	0.8036	0.8015	0.8252	0.8284	0.8262	0.8304	0.8335	0.8314	0.9252	0.9312	0.9274	0.9091	0.9091	0.9	1.0	1.0	1.0	0.0	0.0	0.0
0.0033	5.1669	100000	0.2070	19.5727	0.8112	0.8105	0.8103	0.8356	0.8350	0.8347	0.8410	0.8407	0.8403	0.9364	0.9383	0.9366	0.9091	0.9091	0.9091	1.0	1.0	1.0	0.1429	0.1111	0.125
0.0058	5.6836	110000	0.2166	19.9157	0.8124	0.8128	0.8121	0.8371	0.8377	0.8368	0.8426	0.8434	0.8424	0.9334	0.9368	0.9344	0.9091	0.9091	0.9091	1.0	1.0	1.0	0.1429	0.1	0.1176
0.0022	6.2003	120000	0.2149	19.3398	0.8147	0.8156	0.8146	0.8385	0.8395	0.8384	0.8438	0.8450	0.8438	0.9363	0.9395	0.9371	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.1429	0.1111	0.125
0.0032	6.7170	130000	0.2221	19.3744	0.8092	0.8134	0.8107	0.8333	0.8378	0.8350	0.8392	0.8439	0.8409	0.9326	0.9391	0.9351	0.9091	0.9167	0.9091	1.0	1.0	1.0	0.0909	0.0667	0.0769
0.0033	7.2336	140000	0.2266	19.7017	0.8094	0.8137	0.8109	0.8344	0.8390	0.8361	0.8397	0.8447	0.8416	0.9317	0.9391	0.9347	0.9091	0.9167	0.9091	1.0	1.0	1.0	0.0	0.0	0.0
0.0028	7.7503	150000	0.2305	19.2454	0.8130	0.8163	0.8141	0.8365	0.8400	0.8377	0.8414	0.8456	0.8430	0.9331	0.9380	0.9348	0.9091	0.9167	0.9091	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0018	8.2670	160000	0.2335	19.0220	0.8196	0.8232	0.8209	0.8429	0.8468	0.8443	0.8480	0.8524	0.8496	0.9366	0.9435	0.9393	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.1429	0.125	0.1333
0.005	8.7837	170000	0.2321	18.8961	0.8197	0.8217	0.8202	0.8432	0.8453	0.8437	0.8482	0.8508	0.8489	0.9372	0.9415	0.9387	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.0	0.0	0.0
0.0015	9.3004	180000	0.2320	18.7734	0.8198	0.8223	0.8205	0.8429	0.8456	0.8437	0.8477	0.8508	0.8487	0.9381	0.9431	0.9399	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.0	0.0	0.0
0.0032	9.8171	190000	0.2357	18.4178	0.8207	0.8226	0.8211	0.8440	0.8460	0.8445	0.8485	0.8513	0.8493	0.9390	0.9434	0.9405	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.0	0.0	0.0
0.0025	10.3338	200000	0.2467	19.3430	0.8115	0.8124	0.8114	0.8346	0.8355	0.8345	0.8395	0.8409	0.8397	0.9325	0.9363	0.9337	0.9091	0.9091	0.9091	1.0	1.0	1.0	0.0	0.0	0.0
0.0007	10.8505	210000	0.2415	18.6381	0.8175	0.8200	0.8182	0.8411	0.8437	0.8419	0.8455	0.8484	0.8464	0.9381	0.9430	0.9399	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.0	0.0	0.0
0.0009	11.3672	220000	0.2405	18.0843	0.8247	0.8244	0.8241	0.8475	0.8474	0.8469	0.8522	0.8526	0.8519	0.9410	0.9437	0.9416	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.1	0.1	0.1000
0.0027	11.8838	230000	0.2407	18.3517	0.8176	0.8208	0.8187	0.8405	0.8438	0.8416	0.8455	0.8491	0.8468	0.9377	0.9438	0.9401	0.9167	0.9167	0.9231	1.0	1.0	1.0	0.0	0.0	0.0
0.0004	12.4005	240000	0.2424	18.4808	0.8130	0.8141	0.8130	0.8361	0.8374	0.8362	0.8412	0.8428	0.8415	0.9373	0.9409	0.9384	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.0909	0.0833	0.0870
0.0006	12.9172	250000	0.2495	18.2982	0.8213	0.8236	0.8219	0.8454	0.8480	0.8461	0.8501	0.8533	0.8512	0.9374	0.9428	0.9394	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.0	0.0	0.0
0.0015	13.4339	260000	0.2507	17.9647	0.8283	0.8315	0.8294	0.8508	0.8541	0.8519	0.8554	0.8589	0.8566	0.9401	0.9460	0.9424	0.9167	0.9231	0.9231	1.0	1.0	1.0	0.1	0.1	0.1000
0.0005	13.9506	270000	0.2497	18.2668	0.8255	0.8255	0.8250	0.8481	0.8483	0.8477	0.8528	0.8534	0.8526	0.9387	0.9418	0.9396	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.0625	0.0714	0.0667
0.0003	14.4673	280000	0.2543	18.2038	0.8243	0.8289	0.8261	0.8466	0.8515	0.8485	0.8513	0.8567	0.8534	0.9359	0.9427	0.9386	0.9167	0.9167	0.9167	1.0	1.0	1.0	0.0	0.0	0.0
0.0004	14.9840	290000	0.2506	17.8357	0.8278	0.8286	0.8277	0.8503	0.8513	0.8503	0.8552	0.8568	0.8555	0.9404	0.9445	0.9418	0.9167	0.9231	0.9231	1.0	1.0	1.0	0.0909	0.0909	0.0909
0.0005	15.5007	300000	0.2551	17.8923	0.8219	0.8259	0.8234	0.8449	0.8492	0.8465	0.8495	0.8541	0.8513	0.9391	0.9453	0.9415	0.9167	0.9167	0.9231	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0003	16.0174	310000	0.2566	17.6469	0.8287	0.8285	0.8281	0.8517	0.8516	0.8512	0.8567	0.8568	0.8562	0.9431	0.9454	0.9437	0.9231	0.9167	0.9231	1.0	1.0	1.0	0.0769	0.1	0.0870
0.0005	16.5340	320000	0.2564	17.8168	0.8242	0.8279	0.8255	0.8478	0.8516	0.8492	0.8525	0.8564	0.8539	0.9428	0.9477	0.9446	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.1	0.1176
0.0003	17.0507	330000	0.2518	17.5682	0.8289	0.8328	0.8304	0.8521	0.8562	0.8536	0.8565	0.8608	0.8581	0.9439	0.9501	0.9464	0.9167	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0002	17.5674	340000	0.2561	17.6311	0.8281	0.8300	0.8286	0.8508	0.8528	0.8513	0.8555	0.8577	0.8561	0.9434	0.9469	0.9445	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0002	18.0841	350000	0.2566	17.5399	0.8275	0.8301	0.8283	0.8505	0.8532	0.8513	0.8552	0.8582	0.8562	0.9433	0.9487	0.9453	0.9167	0.9231	0.9231	1.0	1.0	1.0	0.0	0.0	0.0
0.0007	18.6008	360000	0.2592	17.6437	0.8299	0.8336	0.8312	0.8530	0.8568	0.8544	0.8574	0.8616	0.8590	0.9432	0.9487	0.9453	0.9167	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0003	19.1175	370000	0.2609	17.4958	0.8337	0.8362	0.8344	0.8570	0.8597	0.8578	0.8615	0.8646	0.8625	0.9450	0.9499	0.9468	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.0	0.0	0.0
0.0	19.6342	380000	0.2601	17.5053	0.8286	0.8305	0.8291	0.8513	0.8534	0.8519	0.8558	0.8582	0.8565	0.9446	0.9486	0.9460	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0	20.1509	390000	0.2595	16.9200	0.8378	0.8387	0.8378	0.8602	0.8610	0.8601	0.8648	0.8658	0.8648	0.9480	0.9499	0.9484	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1	0.1	0.1000
0.0	20.6676	400000	0.2615	17.1528	0.8337	0.8337	0.8332	0.8558	0.8558	0.8553	0.8603	0.8608	0.8601	0.9464	0.9490	0.9470	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.1111	0.125
0.0	21.1843	410000	0.2617	17.0333	0.8351	0.8345	0.8344	0.8565	0.8559	0.8557	0.8608	0.8605	0.8601	0.9472	0.9484	0.9472	0.9231	0.9231	0.9286	1.0	1.0	1.0	0.0	0.0	0.0
0.0	21.7009	420000	0.2644	17.0301	0.8360	0.8353	0.8352	0.8586	0.8578	0.8577	0.8635	0.8631	0.8628	0.9484	0.9498	0.9485	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0002	22.2176	430000	0.2624	17.1591	0.8329	0.8333	0.8326	0.8550	0.8555	0.8548	0.8599	0.8608	0.8598	0.9470	0.9494	0.9475	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.1	0.1176
0.0	22.7343	440000	0.2643	17.1402	0.8330	0.8347	0.8333	0.8549	0.8567	0.8553	0.8597	0.8618	0.8602	0.9455	0.9487	0.9464	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0	23.2510	450000	0.2634	16.8287	0.8358	0.8366	0.8357	0.8576	0.8584	0.8575	0.8622	0.8634	0.8623	0.9465	0.9492	0.9472	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1	0.0714	0.0833
0.0	23.7677	460000	0.2643	16.8098	0.8331	0.8343	0.8332	0.8553	0.8567	0.8555	0.8600	0.8618	0.8604	0.9474	0.9507	0.9485	0.9231	0.9231	0.9231	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0	24.2844	470000	0.2641	16.9294	0.8361	0.8372	0.8362	0.8581	0.8592	0.8581	0.8626	0.8643	0.8629	0.9471	0.9505	0.9482	0.9231	0.9231	0.9286	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0	24.8011	480000	0.2646	16.8004	0.8338	0.8348	0.8338	0.8553	0.8563	0.8553	0.8600	0.8617	0.8603	0.9465	0.9497	0.9475	0.9231	0.9231	0.9286	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0	25.3178	490000	0.2639	16.7154	0.8372	0.8387	0.8374	0.8589	0.8605	0.8592	0.8636	0.8658	0.8642	0.9475	0.9509	0.9486	0.9231	0.9231	0.9286	1.0	1.0	1.0	0.1429	0.125	0.1333
0.0	25.8345	500000	0.2642	16.6210	0.8369	0.8382	0.8371	0.8586	0.8599	0.8588	0.8633	0.8652	0.8637	0.9480	0.9510	0.9488	0.9231	0.9231	0.9286	1.0	1.0	1.0	0.1429	0.125	0.1333

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
License: apache-2.0
Languages: he
Created: Jul 17, 2024
Updated: Jul 18, 2024