whisper-small-basque-lr1e-5

Ansu/whisper-small-basque-lr1e-5

whisper-small-basque-lr1e-5 is a machine learning model from Ansu. Use it for the machine learning task on the model card, and read the license before you ship it in a product. 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

225

Public

Repo size

9.7 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin967 MB · 100%

At a glance

License: apache-2.0
Model type: whisper
Access: Public
Created: Jul 28, 2026
Updated: Jul 28, 2026
SHA: 64abb52d

Model card

From the Hugging Face model README

whisper-small-basque-lr1e-5-freezeFalse

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

Loss: 0.1860
Wer: 8.6237

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: 128
eval_batch_size: 64
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: 10000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1695	0.33	1000	0.2690	14.8682
0.1145	0.66	2000	0.2183	11.4525
0.094	0.99	3000	0.1978	10.1349
0.0692	1.32	4000	0.1933	9.5979
0.0625	1.65	5000	0.1860	9.0421
0.0607	1.98	6000	0.1859	9.1857
0.0396	2.31	7000	0.1848	8.6487
0.0388	2.64	8000	0.1843	8.6299
0.0391	2.97	9000	0.1830	8.5051
0.0322	3.3	10000	0.1860	8.6237

Framework versions

Transformers 4.25.1
Pytorch 2.5.1+cu121
Datasets 2.8.0
Tokenizers 0.13.3

Type: whisper
License: apache-2.0
Created: Jul 28, 2026
Updated: Jul 28, 2026