geez-asr

This model is a fine-tuned version of Samuael/asr-amharic-phoneme-based-37-6 on the alffa_amharic dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.1275
• eval_wer: 0.1471
• eval_phoneme_cer: 0.0287
• eval_cer: 0.0405
• eval_runtime: 34.6716
• eval_samples_per_second: 18.17
• eval_steps_per_second: 2.279
• epoch: 10.2915
• step: 46600

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
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 20
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.45.0
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.20.0