ronanhansel/wav2vec2-vocoder-ft

ronanhansel/wav2vec2-vocoder

This model is a fine-tuned version of facebook/wav2vec2-base on an LibriSeVoc dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0038
• Train Accuracy: 0.9988
• Validation Loss: 0.0328
• Validation Accuracy: 0.9884
• Epoch: 2

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:

• optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'module': 'transformers.optimization_tf', 'class_name': 'WarmUp', 'config': {'initial_learning_rate': 3e-05, 'decay_schedule_fn': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 414, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 45, 'power': 1.0, 'name': None}, 'registered_name': 'WarmUp'}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
• training_precision: float32

Training results

Train Loss	Train Accuracy	Validation Loss	Validation Accuracy	Epoch
0.0487	0.9837	0.1471	0.9613	0
0.0380	0.9898	0.0253	0.9884	1
0.0038	0.9988	0.0328	0.9884	2

Framework versions

• Transformers 4.51.3
• TensorFlow 2.18.0
• Datasets 3.5.1
• Tokenizers 0.21.1