asr-africa/wav2vec2-xlsr-fula-google-fleurs-5-hours-plus-lm

wav2vec2-xlsr-fula-google-fleurs-5-hours

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

• Loss: 1.1949
• Wer: 0.6460
• Cer: 0.2359

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: 0.0003
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
7.1138	10.96	200	2.9561	1.0	1.0
2.8708	21.92	400	2.0221	1.0	0.6369
1.0031	32.88	600	0.9750	0.6509	0.2222
0.4471	43.84	800	1.1949	0.6460	0.2359

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.0+cu118
• Datasets 2.17.0
• Tokenizers 0.15.2