wav2vec2-xls-r-300m-fr-5min

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: 1.5016
• Wer: 0.6981
• Cer: 0.2705

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: 7e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
2.0858	57.1429	400	1.9368	0.9322	0.5048
0.9206	114.2857	800	1.3725	0.7727	0.3027
0.598	171.4286	1200	1.3869	0.7196	0.2792
0.3647	228.5714	1600	1.4650	0.7033	0.2710
0.3468	285.7143	2000	1.5016	0.6981	0.2705

Framework versions

• Transformers 4.50.0
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1