<!-- 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-0.25-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: 100.6813
- Wer: 0.1288
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 |
---|---|---|---|---|
633.8748 | 0.75 | 100 | 196.0037 | 0.1739 |
464.4784 | 1.49 | 200 | 158.6537 | 0.1839 |
402.8189 | 2.24 | 300 | 133.1326 | 0.1780 |
375.327 | 2.99 | 400 | 124.3412 | 0.1694 |
350.275 | 3.73 | 500 | 119.5099 | 0.1563 |
336.7376 | 4.48 | 600 | 115.2354 | 0.1565 |
322.0894 | 5.22 | 700 | 114.3447 | 0.1480 |
312.4163 | 5.97 | 800 | 112.2822 | 0.1488 |
308.7276 | 6.72 | 900 | 111.4043 | 0.1446 |
305.5383 | 7.46 | 1000 | 109.4352 | 0.1427 |
300.5675 | 8.21 | 1100 | 108.9405 | 0.1397 |
296.3556 | 8.96 | 1200 | 107.5694 | 0.1416 |
294.1826 | 9.7 | 1300 | 106.5605 | 0.1384 |
290.6859 | 10.45 | 1400 | 106.9627 | 0.1365 |
286.1896 | 11.19 | 1500 | 106.2790 | 0.1343 |
282.0118 | 11.94 | 1600 | 105.8356 | 0.1363 |
277.4627 | 12.69 | 1700 | 105.2998 | 0.1354 |
282.7745 | 13.43 | 1800 | 105.2338 | 0.1347 |
275.9889 | 14.18 | 1900 | 104.4767 | 0.1331 |
272.8276 | 14.93 | 2000 | 104.2786 | 0.1346 |
272.7259 | 15.67 | 2100 | 103.5771 | 0.1330 |
267.0375 | 16.42 | 2200 | 102.2268 | 0.1334 |
269.7613 | 17.16 | 2300 | 102.9402 | 0.1316 |
272.2131 | 17.91 | 2400 | 102.2786 | 0.1333 |
269.4063 | 18.66 | 2500 | 101.9838 | 0.1336 |
260.0655 | 19.4 | 2600 | 102.0956 | 0.1305 |
262.16 | 20.15 | 2700 | 101.6337 | 0.1309 |
262.3828 | 20.9 | 2800 | 102.1123 | 0.1302 |
261.6759 | 21.64 | 2900 | 101.4485 | 0.1301 |
264.2934 | 22.39 | 3000 | 101.2778 | 0.1293 |
264.9853 | 23.13 | 3100 | 101.4405 | 0.1298 |
262.5046 | 23.88 | 3200 | 101.3836 | 0.1299 |
263.8816 | 24.63 | 3300 | 101.2221 | 0.1293 |
261.7403 | 25.37 | 3400 | 101.1367 | 0.1280 |
258.1267 | 26.12 | 3500 | 100.7606 | 0.1293 |
255.0854 | 26.87 | 3600 | 100.6364 | 0.1289 |
256.0877 | 27.61 | 3700 | 100.6879 | 0.1287 |
258.1701 | 28.36 | 3800 | 100.6964 | 0.1287 |
256.9546 | 29.1 | 3900 | 100.6813 | 0.1288 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.7.1
- Tokenizers 0.11.0