shook-tiny

b1n1yam/shook-tiny

shook-tiny is a automatic speech recognition model from b1n1yam. 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

3% of all-time downloads

All-time downloads

222

Public

Parameters

37.8M

2.1 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: Nov 13, 2025
Updated: Nov 15, 2025
SHA: 145a4fed

Base models

openai/whisper-tiny

Model card

From the Hugging Face model README

tiny continued from check point 8e-6 - Biniyam Daniel

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: 0.0534
Wer: 18.3821

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: 8e-06
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
training_steps: 5088
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0608	0.0337	100	0.0588	20.9205
0.0617	0.0675	200	0.0568	20.5858
0.0477	0.1012	300	0.0563	20.6137
0.0526	0.1350	400	0.0558	20.1953
0.0555	0.1687	500	0.0560	19.9721
0.0542	0.2025	600	0.0564	20.5858
0.0458	0.2362	700	0.0557	20.8368
0.0439	0.2700	800	0.0559	20.0837
0.0419	0.3037	900	0.0562	20.5021
0.0469	0.3375	1000	0.0556	19.6653
0.0457	0.3712	1100	0.0550	20.0
0.0465	0.4050	1200	0.0550	19.7768
0.0453	0.4387	1300	0.0552	19.3863
0.0425	0.4725	1400	0.0558	19.6095
0.0464	0.5062	1500	0.0547	19.6653
0.0396	0.5400	1600	0.0545	19.3863
0.043	0.5737	1700	0.0551	19.6653
0.0415	0.6075	1800	0.0550	19.3305
0.0396	0.6412	1900	0.0546	18.4937
0.0409	0.6750	2000	0.0542	18.7448
0.0418	0.7087	2100	0.0534	19.0237
0.0446	0.7425	2200	0.0538	19.1074
0.0364	0.7762	2300	0.0537	18.6053
0.0343	0.8100	2400	0.0537	18.4658
0.0437	0.8437	2500	0.0532	18.4100
0.0386	0.8775	2600	0.0530	18.9121
0.0426	0.9112	2700	0.0534	18.2706
0.0372	0.9450	2800	0.0536	18.6890
0.0325	0.9787	2900	0.0533	18.5495
0.03	1.0125	3000	0.0537	18.4100
0.0253	1.0462	3100	0.0545	18.5774
0.0316	1.0800	3200	0.0550	18.4658
0.0251	1.1137	3300	0.0556	18.7727
0.0261	1.1475	3400	0.0554	18.2427
0.0285	1.1812	3500	0.0551	18.4658
0.0234	1.2150	3600	0.0553	18.6890
0.0369	1.2487	3700	0.0549	18.3543
0.0248	1.2825	3800	0.0553	18.2985
0.0238	1.3162	3900	0.0551	18.2985
0.0278	1.3500	4000	0.0551	18.1311
0.0351	1.3837	4100	0.0544	18.4379
0.0459	1.4175	4200	0.0539	17.9916
0.0469	1.4512	4300	0.0537	18.3543
0.0384	1.4850	4400	0.0536	18.4658
0.0503	1.5187	4500	0.0536	18.4100
0.0369	1.5525	4600	0.0536	18.2985
0.0373	1.5862	4700	0.0535	18.4100
0.0376	1.6200	4800	0.0534	18.3821
0.0385	1.6537	4900	0.0534	18.3821
0.0362	1.6875	5000	0.0534	18.3821

Framework versions

Transformers 4.57.1
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.22.1

Task: Automatic Speech Recognition
Library: transformers
Type: whisper
License: apache-2.0
Languages: am
Created: Nov 13, 2025
Updated: Nov 15, 2025