All-lang_tag-whisper-lg-3-Nov27

sqrk/All-lang_tag-whisper-lg-3-Nov27

All-lang_tag-whisper-lg-3-Nov27 is a machine learning model from sqrk. 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

40% of all-time downloads

All-time downloads

Public

Parameters

1.5B

18.5 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors6.2 GB · 100%

At a glance

License: apache-2.0
Model type: whisper
Access: Public
Created: Nov 27, 2024
Updated: Nov 28, 2024
SHA: 7a9578fd

Base models

openai/whisper-large-v3

Model card

From the Hugging Face model README

All-lang_tag-whisper-lg-3-Nov27

This model is a fine-tuned version of openai/whisper-large-v3 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.3328
Wer: 36.7173

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.873	0.0266	100	0.7762	59.1708
0.6282	0.0532	200	0.5069	43.8492
0.4476	0.0798	300	0.4048	37.4254
0.3879	0.1065	400	0.3787	37.6813
0.3497	0.1331	500	0.3328	36.7173

Framework versions

Transformers 4.43.4
Pytorch 2.4.1
Datasets 3.0.0
Tokenizers 0.19.1

Type: whisper
License: apache-2.0
Created: Nov 27, 2024
Updated: Nov 28, 2024