trained_french

This model is a fine-tuned version of facebook/wav2vec2-base-960h on the None dataset. It achieves the following results on the evaluation set:

• Loss: 4.8493
• Wer: 1.0

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: 0.003
• train_batch_size: 6
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 12
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 20
• num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
6.2268	5.53	50	4.9813	1.0
5.724	11.11	100	4.8808	1.0
5.629	16.63	150	4.9001	1.0
5.3351	22.21	200	4.8457	1.0
5.2043	27.74	250	4.8386	1.0
5.1709	33.32	300	4.8647	1.0
5.065	38.84	350	4.8574	1.0
5.0685	44.42	400	4.8449	1.0
5.0584	49.95	450	4.8412	1.0
4.9626	55.53	500	4.8493	1.0

Framework versions

• Transformers 4.17.0
• Pytorch 1.11.0+cu113
• Datasets 1.18.3
• Tokenizers 0.12.1