data-augmentation-whitenoise-timit-1155

gazzehamine/data-augmentation-whitenoise-timit-1155

data-augmentation-whitenoise-timit-1155 is a automatic speech recognition model from gazzehamine. 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

8% of all-time downloads

All-time downloads

Public

Repo size

755 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin378 MB · 100%

At a glance

Task: Automatic Speech Recognition
Library: transformers
License: apache-2.0
Model type: wav2vec2
Access: Public
Created: Jul 29, 2022
Updated: Jul 30, 2022
SHA: 45574bd6

Model card

From the Hugging Face model README

data-augmentation-whitenoise-timit-1155

This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.5458
Wer: 0.3324

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1000
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.5204	0.8	500	1.6948	0.9531
0.8435	1.6	1000	0.5367	0.5113
0.4449	2.4	1500	0.4612	0.4528
0.3182	3.21	2000	0.4314	0.4156
0.2328	4.01	2500	0.4250	0.4031
0.1897	4.81	3000	0.4630	0.4023
0.1628	5.61	3500	0.4445	0.3922
0.1472	6.41	4000	0.4452	0.3793
0.1293	7.21	4500	0.4715	0.3847
0.1176	8.01	5000	0.4267	0.3757
0.1023	8.81	5500	0.4494	0.3821
0.092	9.62	6000	0.4501	0.3704
0.0926	10.42	6500	0.4722	0.3643
0.0784	11.22	7000	0.5033	0.3765
0.077	12.02	7500	0.5165	0.3684
0.0704	12.82	8000	0.5138	0.3646
0.0599	13.62	8500	0.5664	0.3674
0.0582	14.42	9000	0.5188	0.3575
0.0526	15.22	9500	0.5605	0.3621
0.0512	16.03	10000	0.5400	0.3585
0.0468	16.83	10500	0.5471	0.3603
0.0445	17.63	11000	0.5168	0.3555
0.0411	18.43	11500	0.5772	0.3542
0.0394	19.23	12000	0.5079	0.3567
0.0354	20.03	12500	0.5427	0.3613
0.0325	20.83	13000	0.5532	0.3572
0.0318	21.63	13500	0.5223	0.3514
0.0269	22.44	14000	0.6002	0.3460
0.028	23.24	14500	0.5591	0.3432
0.0254	24.04	15000	0.5837	0.3432
0.0235	24.84	15500	0.5571	0.3397
0.0223	25.64	16000	0.5470	0.3383
0.0193	26.44	16500	0.5611	0.3367
0.0227	27.24	17000	0.5405	0.3342
0.0183	28.04	17500	0.5205	0.3330
0.017	28.85	18000	0.5512	0.3330
0.0167	29.65	18500	0.5458	0.3324

Framework versions

Transformers 4.17.0
Pytorch 1.12.0+cu113
Datasets 1.18.3
Tokenizers 0.12.1

Task: Automatic Speech Recognition
Library: transformers
Type: wav2vec2
License: apache-2.0
Created: Jul 29, 2022
Updated: Jul 30, 2022