whisper-small-finetune

quinnlue/whisper-small-finetune

whisper-small-finetune is a machine learning model from quinnlue. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as apache-2.0.

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

Downloads · 30 days

4% of all-time downloads

All-time downloads

805

Public

Repo size

8.3 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors3.3 GB · 54%

At a glance

Library: peft
License: apache-2.0
Access: Public
Created: Feb 23, 2026
Updated: Mar 1, 2026
SHA: 25e92ade

Base models

openai/whisper-small.en

Model card

From the Hugging Face model README

whisper-small-finetune

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

Loss: 0.2792
Wer: 0.1809

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: 0.0005
train_batch_size: 28
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 56
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 500
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.8781	0.0897	500	0.4294	0.3054
0.7503	0.1794	1000	0.3863	0.4102
0.7661	0.2690	1500	0.3689	0.3739
0.8083	0.3587	2000	0.3561	0.2300
0.7457	0.4484	2500	0.3490	0.2738
0.7190	0.5381	3000	0.3477	0.3135
0.6632	0.6277	3500	0.3407	0.2436
0.6398	0.7174	4000	0.3394	0.2489
0.8473	0.8071	4500	0.3924	0.2328
0.7675	0.8968	5000	0.3755	0.2320
0.7742	0.9865	5500	0.3517	0.2260
0.6977	1.0760	6000	0.3481	0.2846
0.7014	1.1657	6500	0.3324	0.3328
0.7019	1.2554	7000	0.3227	0.2110
0.6231	1.3451	7500	0.3200	0.3270
0.6501	1.4348	8000	0.3130	0.2428
0.7791	1.5244	8500	0.3020	0.1987
0.5489	1.6141	9000	0.2977	0.2580
0.6107	1.7038	9500	0.2955	0.2528
0.6150	1.7935	10000	0.2935	0.1967
0.5966	1.8831	10500	0.2903	0.1889
0.6118	1.9728	11000	0.2871	0.2472
0.4407	2.0624	11500	0.2847	0.1863
0.3791	2.1521	12000	0.2869	0.1873
0.4199	2.2418	12500	0.2879	0.1882
0.4828	2.3315	13000	0.2830	0.1899
0.3780	2.4211	13500	0.2835	0.1859
0.3910	2.5108	14000	0.2820	0.1837
0.3888	2.6005	14500	0.2807	0.1855
0.4063	2.6902	15000	0.2801	0.1834
0.4369	2.7798	15500	0.2795	0.1813
0.4117	2.8695	16000	0.2793	0.1811
0.3622	2.9592	16500	0.2793	0.1810

Framework versions

PEFT 0.18.1
Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

Library: peft
License: apache-2.0
Created: Feb 23, 2026
Updated: Mar 1, 2026