metadata

license: apache-2.0
base_model: facebook/wav2vec2-xls-r-300m
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: mascir_fr_wav2vec_version1000
    results: []

mascir_fr_wav2vec_version1000

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.4441
Wer: 0.3622

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	2.0	250	4.6558	1.0
5.4653	4.0	500	3.1189	1.0
5.4653	6.0	750	1.3807	0.9344
1.6415	8.0	1000	0.6832	0.5689
1.6415	10.0	1250	0.4986	0.48
0.3065	12.0	1500	0.4968	0.4711
0.3065	14.0	1750	0.4470	0.4533
0.1441	16.0	2000	0.4832	0.4433
0.1441	18.0	2250	0.5433	0.45
0.0938	20.0	2500	0.4734	0.4344
0.0938	22.0	2750	0.4745	0.4111
0.0727	24.0	3000	0.4236	0.4044
0.0727	26.0	3250	0.4692	0.4133
0.0556	28.0	3500	0.4411	0.3967
0.0556	30.0	3750	0.4722	0.3822
0.0422	32.0	4000	0.4845	0.3978
0.0422	34.0	4250	0.4818	0.4
0.0325	36.0	4500	0.4638	0.3944
0.0325	38.0	4750	0.4737	0.38
0.0284	40.0	5000	0.4615	0.3822
0.0284	42.0	5250	0.4491	0.3722
0.0235	44.0	5500	0.4480	0.3744
0.0235	46.0	5750	0.4630	0.3711
0.0172	48.0	6000	0.4421	0.3644
0.0172	50.0	6250	0.4441	0.3622

Framework versions

Transformers 4.31.0
Pytorch 2.0.1+cu118
Datasets 2.14.2
Tokenizers 0.13.3