whisper-finetuned-local

MNG-4/whisper-finetuned-local

whisper-finetuned-local is a automatic speech recognition model from MNG-4. 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

435

Public

Parameters

37.8M

302 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors151 MB · 100%

At a glance

Task: Automatic Speech Recognition
Library: transformers
License: apache-2.0
Model type: whisper
Access: Public
Created: Jan 9, 2026
Updated: Jan 14, 2026
SHA: a545408a

Base models

openai/whisper-tiny

Model card

From the Hugging Face model README

whisper-finetuned-local

This model is a fine-tuned version of openai/whisper-tiny on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.0345
Wer Kik: 0.689
Wer Mean: 0.689
Cer Kik: 0.33
Cer Mean: 0.33

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: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 50
training_steps: 2000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Kik	Wer Mean	Cer Kik	Cer Mean
3.2486	0.7407	50	2.5295	1.051	1.051	0.546	0.546
1.7584	1.4741	100	1.6429	0.851	0.851	0.405	0.405
1.3251	2.2074	150	1.2782	0.797	0.797	0.373	0.373
1.0064	2.9481	200	1.1429	0.76	0.76	0.353	0.353
0.9269	3.6815	250	1.0942	0.741	0.741	0.346	0.346
0.7781	4.4148	300	1.0459	0.73	0.73	0.341	0.341
0.6803	5.1481	350	1.0239	0.712	0.712	0.337	0.337
0.6518	5.8889	400	1.0077	0.702	0.702	0.331	0.331
0.5948	6.6222	450	0.9980	0.699	0.699	0.334	0.334
0.5605	7.3556	500	0.9975	0.702	0.702	0.333	0.333
0.5275	8.0889	550	0.9947	0.693	0.693	0.325	0.325
0.4221	8.8296	600	0.9988	0.697	0.697	0.33	0.33
0.3822	9.5630	650	1.0090	0.693	0.693	0.329	0.329
0.3408	10.2963	700	1.0246	0.7	0.7	0.335	0.335
0.3343	11.0296	750	1.0181	0.691	0.691	0.331	0.331
0.296	11.7704	800	1.0345	0.689	0.689	0.33	0.33

Framework versions

Transformers 4.57.3
Pytorch 2.9.0+cu126
Datasets 4.0.0
Tokenizers 0.22.2

Task: Automatic Speech Recognition
Library: transformers
Type: whisper
License: apache-2.0
Created: Jan 9, 2026
Updated: Jan 14, 2026