whisper-tiny-fr

personalizedrefrigerator/whisper-tiny-fr

whisper-tiny-fr is a automatic speech recognition model from personalizedrefrigerator. 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

26% of all-time downloads

All-time downloads

Public

Parameters

37.8M

1.9 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin229 MB · 60%

At a glance

Task: Automatic Speech Recognition
Library: transformers
License: apache-2.0
Model type: whisper
Access: Public
Created: Dec 3, 2024
Updated: Mar 19, 2025
SHA: 23876a7a

Base models

openai/whisper-tiny

Model card

From the Hugging Face model README

whisper-tiny-fr

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: 0.6094
Model Preparation Time: 0.0026
Wer: 0.3301
Cer: 0.1774

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: 0.0001
train_batch_size: 8
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: linear
training_steps: 33000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Wer	Cer
0.8933	0.0303	1000	1.3184	0.0026	0.5726	0.3162
0.7455	0.0606	2000	1.1738	0.0026	0.5533	0.2930
0.735	0.0909	3000	1.0462	0.0026	0.6032	0.3290
0.6426	0.1212	4000	1.0629	0.0026	0.4937	0.2473
0.6389	0.1515	5000	1.0132	0.0026	0.5671	0.3415
0.532	0.1818	6000	1.0194	0.0026	0.4639	0.2355
0.5341	0.2121	7000	1.0092	0.0026	0.4619	0.2435
0.4773	0.2424	8000	0.9672	0.0026	0.4666	0.2822
0.4932	0.2727	9000	0.9778	0.0026	0.4175	0.2187
0.479	0.3030	10000	0.9639	0.0026	0.4105	0.2169
0.4663	0.3333	11000	0.9689	0.0026	0.4236	0.2245
0.3647	0.3636	12000	1.0025	0.0026	0.4326	0.2297
0.451	0.3939	13000	0.8810	0.0026	0.4648	0.2591
0.4522	0.4242	14000	0.8283	0.0026	0.3869	0.1965
0.5064	0.4545	15000	0.8165	0.0026	0.3703	0.1898
0.4355	0.4848	16000	0.7857	0.0026	0.4367	0.2257
0.2953	0.5152	17000	0.8007	0.0026	0.3650	0.2020
0.4345	0.5455	18000	0.7823	0.0026	0.4544	0.2381
0.4117	0.5758	19000	0.7648	0.0026	0.3595	0.1823
0.4071	0.6061	20000	0.7475	0.0026	0.4121	0.2049
0.4371	0.6364	21000	0.7285	0.0026	0.3509	0.1842
0.34	0.6667	22000	0.7686	0.0026	0.3566	0.1860
0.335	0.6970	23000	0.7514	0.0026	0.3595	0.1846
0.2946	0.7273	24000	0.7928	0.0026	0.3742	0.2006
0.3916	0.7576	25000	0.6843	0.0026	0.3416	0.1747
0.3233	0.7879	26000	0.6478	0.0026	0.3178	0.1626
0.2981	0.8182	27000	0.6737	0.0026	0.3274	0.1669
0.2945	0.8485	28000	0.6512	0.0026	0.3302	0.1643
0.2956	0.8788	29000	0.6867	0.0026	0.3925	0.1991
0.2541	0.9091	30000	0.6333	0.0026	0.3186	0.1620
0.2475	0.9394	31000	0.7018	0.0026	0.3343	0.1722
0.2951	0.9697	32000	0.6527	0.0026	0.3168	0.1632
0.2879	1.0	33000	0.6440	0.0026	0.3102	0.1560

Framework versions

Transformers 4.49.0
Pytorch 2.6.0+cu124
Datasets 3.4.1
Tokenizers 0.21.1

Task: Automatic Speech Recognition
Library: transformers
Type: whisper
License: apache-2.0
Created: Dec 3, 2024
Updated: Mar 19, 2025