ylacombe/wav2vec2-bert-CV16-en-cv-2
112
<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
wav2vec2-bert-CV16-en-cv-2
This model is a fine-tuned version of facebook/w2v-bert-2.0 on the MOZILLA-FOUNDATION/COMMONVOICE16_0 - EN dataset. It achieves the following results on the evaluation set:
- Loss: 0.4363
- Wer: 0.1304
- Cer: 0.0527
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: 3e-05
- trainbatchsize: 12
- evalbatchsize: 12
- seed: 42
- distributed_type: multi-GPU
- num_devices: 3
- gradientaccumulationsteps: 3
- totaltrainbatch_size: 108
- totalevalbatch_size: 36
- optimizer: Adam with betas=(0.9,0.96) and epsilon=1e-08
- lrschedulertype: linear
- lrschedulerwarmup_steps: 10000
- num_epochs: 50.0
- mixedprecisiontraining: Native AMP
Training results
Framework versions
- Transformers 4.37.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
