metadata
license: apache-2.0
tags:
- automatic-speech-recognition
- gary109/AI_Light_Dance
- generated_from_trainer
model-index:
- name: ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v2
results: []
ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v2
This model is a fine-tuned version of gary109/ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1 on the GARY109/AI_LIGHT_DANCE - ONSET-SINGING3 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5105
- Wer: 0.2552
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: 5e-06
- 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: 50
- num_epochs: 50.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.6154 | 1.0 | 72 | 0.5266 | 0.2551 |
| 0.5958 | 2.0 | 144 | 0.5272 | 0.2586 |
| 0.5825 | 3.0 | 216 | 0.5249 | 0.2587 |
| 0.5717 | 4.0 | 288 | 0.5236 | 0.2571 |
| 0.5831 | 5.0 | 360 | 0.5203 | 0.2590 |
| 0.5652 | 6.0 | 432 | 0.5127 | 0.2575 |
| 0.5665 | 7.0 | 504 | 0.5229 | 0.2587 |
| 0.5625 | 8.0 | 576 | 0.5248 | 0.2547 |
| 0.5661 | 9.0 | 648 | 0.5214 | 0.2558 |
| 0.5583 | 10.0 | 720 | 0.5197 | 0.2582 |
| 0.5605 | 11.0 | 792 | 0.5213 | 0.2611 |
| 0.5784 | 12.0 | 864 | 0.5328 | 0.2583 |
| 0.5636 | 13.0 | 936 | 0.5246 | 0.2586 |
| 0.5581 | 14.0 | 1008 | 0.5230 | 0.2546 |
| 0.567 | 15.0 | 1080 | 0.5205 | 0.2572 |
| 0.5586 | 16.0 | 1152 | 0.5259 | 0.2556 |
| 0.5358 | 17.0 | 1224 | 0.5334 | 0.2605 |
| 0.5526 | 18.0 | 1296 | 0.5181 | 0.2556 |
| 0.5483 | 19.0 | 1368 | 0.5131 | 0.2562 |
| 0.5487 | 20.0 | 1440 | 0.5179 | 0.2561 |
| 0.5489 | 21.0 | 1512 | 0.5259 | 0.2596 |
| 0.5582 | 22.0 | 1584 | 0.5199 | 0.2551 |
| 0.5351 | 23.0 | 1656 | 0.5283 | 0.2535 |
| 0.5572 | 24.0 | 1728 | 0.5120 | 0.2533 |
| 0.5467 | 25.0 | 1800 | 0.5176 | 0.2578 |
| 0.5424 | 26.0 | 1872 | 0.5105 | 0.2552 |
| 0.5344 | 27.0 | 1944 | 0.5212 | 0.2541 |
| 0.5444 | 28.0 | 2016 | 0.5155 | 0.2556 |
| 0.5276 | 29.0 | 2088 | 0.5231 | 0.2551 |
| 0.5501 | 30.0 | 2160 | 0.5224 | 0.2557 |
| 0.5335 | 31.0 | 2232 | 0.5279 | 0.2550 |
| 0.5315 | 32.0 | 2304 | 0.5151 | 0.2545 |
| 0.5344 | 33.0 | 2376 | 0.5204 | 0.2528 |
| 0.5249 | 34.0 | 2448 | 0.5153 | 0.2543 |
| 0.5478 | 35.0 | 2520 | 0.5154 | 0.2544 |
| 0.5346 | 36.0 | 2592 | 0.5123 | 0.2534 |
| 0.5436 | 37.0 | 2664 | 0.5210 | 0.2565 |
| 0.5299 | 38.0 | 2736 | 0.5182 | 0.2537 |
| 0.5248 | 39.0 | 2808 | 0.5240 | 0.2529 |
| 0.5295 | 40.0 | 2880 | 0.5250 | 0.2563 |
| 0.5343 | 41.0 | 2952 | 0.5179 | 0.2536 |
| 0.5255 | 42.0 | 3024 | 0.5213 | 0.2560 |
| 0.525 | 43.0 | 3096 | 0.5221 | 0.2553 |
| 0.5345 | 44.0 | 3168 | 0.5230 | 0.2531 |
| 0.5485 | 45.0 | 3240 | 0.5212 | 0.2537 |
| 0.5471 | 46.0 | 3312 | 0.5215 | 0.2532 |
| 0.5375 | 47.0 | 3384 | 0.5216 | 0.2544 |
| 0.5229 | 48.0 | 3456 | 0.5209 | 0.2551 |
| 0.5218 | 49.0 | 3528 | 0.5216 | 0.2536 |
| 0.5292 | 50.0 | 3600 | 0.5208 | 0.2545 |
Framework versions
- Transformers 4.21.0.dev0
- Pytorch 1.9.1+cu102
- Datasets 2.3.3.dev0
- Tokenizers 0.12.1