wav2vec2-large-xls-r-300m-rw-KinyarwandaTTSDataset-1hr-v1

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: 0.0032
• Wer: 0.0063
• Cer: 0.0012

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: 0.0005
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
4.6667	13.3333	100	2.9117	1.0	1.0
2.7734	26.6667	200	2.0105	0.9985	0.6220
0.6665	40.0	300	0.0973	0.1083	0.0158
0.1884	53.3333	400	0.0365	0.0622	0.0074
0.1423	66.6667	500	0.0127	0.0231	0.0033
0.1042	80.0	600	0.0063	0.0082	0.0020
0.0698	93.3333	700	0.0032	0.0063	0.0012

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1