metadata

license: apache-2.0
tags:
  - automatic-speech-recognition
  - BembaSpeech
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: xls-r-300m-bemba-fullset
    results: []

xls-r-300m-bemba-fullset

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

Loss: 0.2534
Wer: 0.8470

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: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 2000
num_epochs: 2.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.268	0.67	500	1.0355	0.9929
0.7365	1.35	1000	0.2534	0.8470

Framework versions

Transformers 4.25.0.dev0
Pytorch 1.12.1+cu113
Datasets 2.7.0
Tokenizers 0.13.2