metadata
tags:
- automatic-speech-recognition
- gary109/AI_Light_Dance
- generated_from_trainer
model-index:
- name: ai-light-dance_singing2_ft_wav2vec2-large-xlsr-53-5gram-v4-1
results: []
ai-light-dance_singing2_ft_wav2vec2-large-xlsr-53-5gram-v4-1
This model is a fine-tuned version of gary109/ai-light-dance_singing_ft_wav2vec2-large-xlsr-53-5gram-v4 on the GARY109/AI_LIGHT_DANCE - ONSET-SINGING2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2143
- Wer: 0.1211
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
- gradient_accumulation_steps: 8
- 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: 100
- num_epochs: 10.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2609 | 1.0 | 280 | 0.2313 | 0.1376 |
| 0.2297 | 2.0 | 560 | 0.2240 | 0.1397 |
| 0.1951 | 3.0 | 840 | 0.2280 | 0.1361 |
| 0.1816 | 4.0 | 1120 | 0.2215 | 0.1282 |
| 0.1634 | 5.0 | 1400 | 0.2180 | 0.1240 |
| 0.1338 | 6.0 | 1680 | 0.2226 | 0.1241 |
| 0.1411 | 7.0 | 1960 | 0.2143 | 0.1211 |
| 0.1143 | 8.0 | 2240 | 0.2181 | 0.1174 |
| 0.1127 | 9.0 | 2520 | 0.2215 | 0.1167 |
| 0.105 | 10.0 | 2800 | 0.2196 | 0.1160 |
Framework versions
- Transformers 4.21.0.dev0
- Pytorch 1.9.1+cu102
- Datasets 2.3.3.dev0
- Tokenizers 0.12.1