<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
wav2vec2-base-timit-demo-google-colab
This model is a fine-tuned version of facebook/wav2vec2-base-100h on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1428
- Wer: 0.1265
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: 0.0001
- 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: 1000
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
6.924 | 0.26 | 200 | 4.0579 | 1.0 |
2.9378 | 0.51 | 400 | 1.9506 | 0.9319 |
1.4334 | 0.77 | 600 | 0.7472 | 0.4723 |
0.8152 | 1.03 | 800 | 0.5167 | 0.3405 |
0.6369 | 1.28 | 1000 | 0.3825 | 0.2747 |
0.4931 | 1.54 | 1200 | 0.3361 | 0.2407 |
0.4986 | 1.8 | 1400 | 0.3224 | 0.2228 |
0.392 | 2.05 | 1600 | 0.2876 | 0.2086 |
0.3527 | 2.31 | 1800 | 0.3104 | 0.2089 |
0.3171 | 2.57 | 2000 | 0.2431 | 0.1821 |
0.2847 | 2.82 | 2200 | 0.2153 | 0.1776 |
0.3274 | 3.08 | 2400 | 0.2486 | 0.1679 |
0.2901 | 3.34 | 2600 | 0.3754 | 0.1627 |
0.2539 | 3.59 | 2800 | 0.2790 | 0.1642 |
0.2427 | 3.85 | 3000 | 0.2485 | 0.1664 |
0.1992 | 4.11 | 3200 | 0.2184 | 0.1574 |
0.2873 | 4.36 | 3400 | 0.1967 | 0.1547 |
0.2037 | 4.62 | 3600 | 0.2289 | 0.1506 |
0.1967 | 4.88 | 3800 | 0.2263 | 0.1506 |
0.2254 | 5.13 | 4000 | 0.1629 | 0.1463 |
0.1808 | 5.39 | 4200 | 0.2015 | 0.1476 |
0.1762 | 5.65 | 4400 | 0.1948 | 0.1456 |
0.1829 | 5.91 | 4600 | 0.1521 | 0.1437 |
0.1934 | 6.16 | 4800 | 0.1638 | 0.1431 |
0.1643 | 6.42 | 5000 | 0.1476 | 0.1435 |
0.1244 | 6.68 | 5200 | 0.1937 | 0.1394 |
0.1615 | 6.93 | 5400 | 0.1508 | 0.1366 |
0.1708 | 7.19 | 5600 | 0.1298 | 0.1348 |
0.1736 | 7.45 | 5800 | 0.1383 | 0.1344 |
0.1429 | 7.7 | 6000 | 0.1711 | 0.1330 |
0.1453 | 7.96 | 6200 | 0.1844 | 0.1302 |
0.1387 | 8.22 | 6400 | 0.3321 | 0.1297 |
0.1259 | 8.47 | 6600 | 0.1617 | 0.1296 |
0.0874 | 8.73 | 6800 | 0.1432 | 0.1270 |
0.1107 | 8.99 | 7000 | 0.1302 | 0.1280 |
0.1205 | 9.24 | 7200 | 0.1461 | 0.1270 |
0.109 | 9.5 | 7400 | 0.1415 | 0.1271 |
0.1117 | 9.76 | 7600 | 0.1428 | 0.1265 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1