<!-- 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. -->
libri-alpha-0.5-Temp-1-processor-change
This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:
- Loss: 91.9750
- Wer: 0.1187
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: 2e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- 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: 500
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
569.0646 | 0.75 | 100 | 175.3549 | 0.1589 |
440.3574 | 1.49 | 200 | 146.3654 | 0.1640 |
398.2328 | 2.24 | 300 | 128.7082 | 0.1562 |
357.5816 | 2.99 | 400 | 117.7871 | 0.1495 |
344.3317 | 3.73 | 500 | 111.0376 | 0.1417 |
331.0486 | 4.48 | 600 | 106.5447 | 0.1398 |
321.4498 | 5.22 | 700 | 105.1187 | 0.1363 |
305.8177 | 5.97 | 800 | 103.2541 | 0.1365 |
304.2076 | 6.72 | 900 | 105.3060 | 0.1385 |
297.746 | 7.46 | 1000 | 101.1069 | 0.1307 |
285.7675 | 8.21 | 1100 | 99.9853 | 0.1303 |
284.6546 | 8.96 | 1200 | 98.5235 | 0.1292 |
281.672 | 9.7 | 1300 | 97.8004 | 0.1295 |
281.0029 | 10.45 | 1400 | 96.9385 | 0.1278 |
283.847 | 11.19 | 1500 | 96.3700 | 0.1275 |
274.4053 | 11.94 | 1600 | 95.9557 | 0.1281 |
271.8855 | 12.69 | 1700 | 95.5764 | 0.1250 |
275.416 | 13.43 | 1800 | 95.0451 | 0.1266 |
267.7354 | 14.18 | 1900 | 94.6620 | 0.1242 |
273.9816 | 14.93 | 2000 | 95.0889 | 0.1241 |
263.9812 | 15.67 | 2100 | 94.4231 | 0.1241 |
258.6033 | 16.42 | 2200 | 93.8011 | 0.1225 |
260.4275 | 17.16 | 2300 | 94.0336 | 0.1210 |
258.7905 | 17.91 | 2400 | 93.4633 | 0.1216 |
255.6817 | 18.66 | 2500 | 93.0448 | 0.1212 |
252.3298 | 19.4 | 2600 | 92.9945 | 0.1216 |
250.5598 | 20.15 | 2700 | 92.9767 | 0.1200 |
249.4384 | 20.9 | 2800 | 93.1555 | 0.1203 |
255.6291 | 21.64 | 2900 | 92.7784 | 0.1208 |
249.5222 | 22.39 | 3000 | 92.5792 | 0.1203 |
250.498 | 23.13 | 3100 | 92.4570 | 0.1205 |
252.2656 | 23.88 | 3200 | 92.3685 | 0.1199 |
248.1438 | 24.63 | 3300 | 92.3731 | 0.1198 |
240.2946 | 25.37 | 3400 | 92.1875 | 0.1192 |
256.2254 | 26.12 | 3500 | 91.9586 | 0.1192 |
248.603 | 26.87 | 3600 | 91.9599 | 0.1191 |
252.9337 | 27.61 | 3700 | 92.1080 | 0.1189 |
250.9757 | 28.36 | 3800 | 92.1051 | 0.1188 |
248.7415 | 29.1 | 3900 | 91.9927 | 0.1187 |
248.7394 | 29.85 | 4000 | 91.9750 | 0.1187 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.7.1
- Tokenizers 0.11.0