unispeech-sat-base-finetuned-common_voice

aydink/unispeech-sat-base-finetuned-common_voice

unispeech-sat-base-finetuned-common_voice is a audio classification model from aydink. Use it for the audio classification task on the model card, and read the license before you ship it in a product. It is set up for transformers.

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

Downloads · 30 days

2% of all-time downloads

All-time downloads

519

Public

Parameters

94.6M

37.8 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors378 MB · 100%

At a glance

Task: Audio Classification
Library: transformers
Model type: unispeech-sat
Access: Public
Created: May 15, 2024
Updated: May 27, 2024
SHA: e6334ba3

Base models

microsoft/unispeech-sat-base

Model card

From the Hugging Face model README

unispeech-sat-base-finetuned-common_voice

This model is a fine-tuned version of microsoft/unispeech-sat-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0641
Accuracy: 0.9875
F1: 0.9875
Recall: 0.9875
Precision: 0.9878
Mcc: 0.9844
Auc: 0.9999

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
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision	Mcc	Auc
0.2181	1.0	200	0.0440	0.9925	0.9925	0.9925	0.9926	0.9906	0.9981
0.0083	2.0	400	0.0609	0.9875	0.9875	0.9875	0.9880	0.9845	0.9987
0.0035	3.0	600	0.0888	0.98	0.9799	0.9800	0.9806	0.9752	0.9991
0.2407	4.0	800	0.1593	0.9725	0.9726	0.9725	0.9740	0.9660	0.9997
0.0859	5.0	1000	0.1234	0.9775	0.9777	0.9775	0.9790	0.9722	0.9999
0.2073	6.0	1200	0.0851	0.9825	0.9826	0.9825	0.9832	0.9783	0.9999
0.0036	7.0	1400	0.0550	0.9925	0.9925	0.9925	0.9927	0.9907	0.9999
0.0036	8.0	1600	0.0600	0.9925	0.9925	0.9925	0.9927	0.9907	1.0000
0.0013	9.0	1800	0.0645	0.99	0.9900	0.99	0.9903	0.9876	1.0000
0.0048	10.0	2000	0.0641	0.9875	0.9875	0.9875	0.9878	0.9844	0.9999

Framework versions

Transformers 4.41.0
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1

Task: Audio Classification
Library: transformers
Type: unispeech-sat
Created: May 15, 2024
Updated: May 27, 2024